WO2024069848A1 - Control for using edge servers in network services - Google Patents

Abstract

A communication system according to the present disclosure executes a notification process, a reception process, and a transfer process. In the notification process, when a packet addressed to a prescribed destination outside of a packet-relaying communication network is detected in the communication network, the prescribed destination is notified regarding an identifier of a communication device that transmitted the packet. In the reception process, an offload notification for causing at least some of services of the prescribed destination regarding the communication device to be offloaded is received from the prescribed destination. In the transfer process, a packet that is transmitted from the communication device to be offloaded as designated by the offload notification, and that is addressed to the prescribed destination, is transferred to an edge server that is located in the communication network and that provides at least some of the services of the prescribed destination.

Description

Control for utilization of edge servers in network services

This disclosure relates to control for the use of edge servers in network services.

In recent years, technologies that provide users with various entertainment activities, such as games and shopping, in a virtual space have come into widespread use.
Patent Literature 1 discloses a game system that realizes a multiplayer game in which multiple users can participate. In this multiplayer game, player characters corresponding to the multiple users act in the same virtual space and cooperate with each other to accomplish missions.

JP 2022-41645 A

Network services in which various users can enjoy the service together, such as the conventional game system described above, are provided by a server device connected to the Internet, but if a large number of users log in to the service at the same time, the processing load on the server device increases.

The objective of this disclosure is to reduce the processing load on servers that provide network services.

In order to solve the above problem, a communication system according to one aspect of the present disclosure includes one or more processors. At least one of the one or more processors executes a notification process, a reception process, and a forwarding process. The notification process is a process of notifying the specified destination of an identifier of a communication device that transmitted a packet when a packet addressed to a specified destination outside the communication network is detected in a communication network that relays packets. The reception process is a process of receiving an offload notification from the specified destination for offloading at least a part of the service of the specified destination to the communication device. The forwarding process is a process of forwarding a packet addressed to the specified destination, from the communication device to be offloaded specified by the offload notification as a source, to an edge server disposed in the communication network that provides at least a part of the service of the specified destination.

In order to solve the above problem, a communication control method according to one aspect of the present disclosure includes, when a packet addressed to a specific destination outside a communication network is detected in a communication network that relays packets, notifying the specific destination of an identifier of a communication device that transmitted the packet, receiving an offload notification from the specific destination for offloading at least a portion of the services of the specific destination to the communication device, and forwarding the packet addressed to the specific destination, with the communication device designated by the offload notification as the offload target as the source, to an edge server disposed in the communication network that provides at least a portion of the services of the specific destination.

According to one aspect of the present disclosure, the processing load on the server that provides the network service can be reduced.

FIG. 1 is a diagram showing an example of a network configuration of a communication system according to the present embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of the management apparatus. FIG. 3 illustrates an example of the functional configuration of the management device. FIG. 4 shows an example of the functional configuration of the MEC server. FIG. 5 shows an example of the functional configuration of a cloud server. FIG. 6 shows an example of the virtual game space. FIG. 7 is a diagram illustrating the offloading control. FIG. 8 is a diagram for explaining a method for determining an offload destination MEC server. FIG. 9 is a diagram showing an example of the configuration of a ring network. FIG. 10 is a diagram illustrating a change in network slice. FIG. 11 is a diagram illustrating packet control by the UPF. FIG. 12 is a communication sequence diagram according to this embodiment.

Below, an embodiment of the present disclosure will be described in detail with reference to the attached drawings. Among the components disclosed below, those having the same functions will be given the same reference numerals, and their description will be omitted. Note that the embodiment disclosed below is one form of the present disclosure, and should be appropriately modified or changed depending on the configuration of the device and various conditions, and is not limited to the following embodiment. Furthermore, not all of the combinations of features described in this embodiment are necessarily essential means for solving the above problem.

In this embodiment, the fifth generation (5G) mobile network standardized by 3GPP (Third Generation Partnership Project) (registered trademark) is assumed as the network to which the technology disclosed herein is applied. Note that the technology disclosed herein may also be applied to networks other than 5G mobile networks.

Furthermore, in this embodiment, an example is described in which a virtual reality (VR) space on the Internet in which a real user (person) acts as an avatar, or the metaverse, which is the mechanism for such a space, is used. In the following description, the term virtual reality space (or virtual space) refers to a virtual space, such as the metaverse, in which a user can act as an avatar of the user.

The term "connection" used in this description means a logical connection for communication. For example, "B connected to A" means that A and B are logically connected so that they can communicate. A and B do not need to be directly physically connected by a physical cable or the like, and there may be multiple devices or wireless communication between A and B.

<Network configuration>
FIG. 1 is a diagram showing an example of a network configuration of a communication system 1 according to the present embodiment.
The communication system 1 includes user equipment (UE) 10a to 10d, a communication network 20, a cloud 30, and a cloud server 40. The communication network 20 includes a base station 21, an MEC server 22, and a management device 23.

The communication network 20 may be a mobile network that relays communications (packets) from the UEs 10a, 10b to their destination.
The mobile network 20 may be, for example, a mobile network in which one MNO (Mobile Network Operator) performs end-to-end network management. The mobile network 20 has a radio access network (RAN) to which the UEs 10a and 10b directly access, and a core network that aggregates multiple RANs.

In the mobile network 20, UEs 10a and 10b communicate wirelessly with the RAN, and the information is sent to the core network for processing. UEs 10a and 10b can use network services (cloud services) 100 provided by a cloud server 40 present on a cloud 30 via the mobile network 20. UEs 10a and 10b can also connect to other companies' mobile networks 50 and 60 via the mobile network 20 to make voice calls with other UEs 10c and 10d.

Here, the UEs 10a to 10d may be communication terminals (communication devices) capable of mobile communication, such as smartphones, tablet terminals, personal computers (PCs), etc. In addition, the UEs 10a to 10d may be wearable terminals, such as head-mounted displays (HMDs) and smart glasses, stationary terminals, such as desktop PCs, game controllers, etc.
Each of the UEs 10a to 10d has a display unit such as a liquid crystal display, and each user can perform various operations using a GUI (Graphic User Interface) provided on the display unit. Note that each of the UEs 10a to 10d may have a separate display unit.
In this embodiment, UEs 10a and 10b are UEs used by users (subscribers) who have entered into a line contract with the MNO that manages the mobile network 20, and UEs 10c and 10d are UEs used by users who are not the above-mentioned subscribers.

The UEs 10a and 10b can access the cloud service 100 via the mobile network 20, the UE 10c via the mobile network 50, and the UE 10d via the mobile network 60.
Furthermore, the UEs 10a to 10d can also connect to the cloud 30 via another network, such as Wi-Fi (registered trademark). The two-dot chain line in Fig. 1 indicates a case where the UE 10b connects to the cloud 30 via another network. In this manner, the UEs 10a to 10d can access the cloud service 100 via various networks.
Here, the cloud 30 refers to a server or system outside the mobile network 20, and may be the Internet or a data center connected to the Internet.
It should be noted that the number of UEs that can be connected to the cloud 30 is not limited to the number shown in Fig. 1. In addition, in the following description, the UEs that can be connected to the cloud 30 may be collectively referred to as UEs 10.

The mobile network 20 can accommodate a plurality of base stations 21. The base station 21 includes an antenna, a distribution board, a battery, etc. The base station 21 performs the function of a Radio Unit (RU), which is part of the function of the RAN.
The mobile network 20 in this embodiment may be a virtualized network built on a virtualization platform. In this case, the mobile network 20 can realize the functions from the backbone network switch to the base station wireless access function by software on a general-purpose server.

At least one MEC server 22 is deployed in the mobile network 20. The MEC server 22 is a server device (edge server) for multi-access edge computing (MEC).
In the present embodiment, the MEC server 22 is disposed in the core network of the mobile network 20, but the base station 21 may be configured to have the function of the MEC server. In addition, although only one MEC server 22 is shown in Fig. 1, the number of MEC servers is not particularly limited.

The management device 23 is placed in the mobile network 20, and performs offloading control in cooperation with the cloud server 40. Here, offloading control refers to control in which an agent executes processing in place of the original executor. In this embodiment, at least a part of the processing executed in the cloud 30 is executed by the MEC server 22. This reduces the load on the cloud 30 (offloading). Also, changing the executor of processing for a UE 10 is simply described as offloading the UE 10. The management device 23 performs offloading control to offload a specific UE 10 to the MEC server 22 according to a notification from the cloud server 40. The offloading control will be described later.

The cloud server 40 is located on the cloud 30 and provides a cloud service 100, which is an external service outside the mobile network 20. The cloud service 100 may be a service that provides VR content that constitutes a virtual space (metaverse) incorporating functions such as avatars, games, communication, shopping, and advertising. The VR content includes at least VR images, which are image data that constitute the virtual space. The VR content may be expressed by images (still images and/or videos), documents, audio, music, or a combination of any two or more of these elements.

The cloud server 40 may include a content database that stores data for generating VR content. The cloud server 40 can acquire information (user operation information) indicating an operation input by a user to the UE 10, generate VR content using data stored in the content database based on the user operation information, and deliver the VR content to the UE 10.
The UE 10 outputs (displays) the VR content, and the user can enjoy the virtual space by viewing the VR content.

In this embodiment, a case will be described in which the cloud service 100 is a service that provides a virtual game space for online games.
The online game may be a multiplayer game in which multiple users can participate. In this online game, multiple avatars A to D corresponding to multiple users respectively can act in the same virtual game space, as shown in Fig. 1. Here, the avatars A to D correspond to users UE 10a to 10d, respectively.

<Hardware Configuration>
FIG. 2 is a diagram illustrating an example of the hardware configuration of the management device 23. As shown in FIG.
2, the management device 23 includes, as an example of a hardware configuration, a CPU 1, a ROM 2, a RAM 3, a HDD 4, a communication I/F 5, and a system bus 6. The management device 23 may also include an external memory (not shown).
The CPU 1 is made up of one or more processors, and performs overall control of the operation of the management device 23. The CPU 1 controls each of the components (2 to 5) via a system bus 6, which is a data transmission path.
The CPU 1 may be replaced by one or more processors such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a graphics processing unit (GPU), or the like.

The ROM 2 is a non-volatile memory that stores control programs and the like necessary for the CPU 1 to execute processing. Note that the programs may be stored in a non-volatile memory such as the HDD 5, or in an external memory such as a removable storage medium (not shown).
The RAM 3 is a volatile memory and functions as a main memory, a work area, etc. of the CPU 1. That is, when executing a process, the CPU 1 loads necessary programs, etc. from the ROM 2 into the RAM 3, and executes the programs, etc. to realize various functional operations.

The HDD 4 stores, for example, various data and various information required when the CPU 1 performs processing using a program. The HDD 4 also stores, for example, various data and various information obtained when the CPU 1 performs processing using a program. Note that the storage may be performed using an external memory such as a non-volatile memory such as an SSD or a removable storage medium together with the HDD 4 or instead of the HDD 4.
The communication I/F (Interface) 5 is an interface that controls communication between the management device 23 and external devices.

The UE 10, the MEC server 22, and the cloud server 40 may have the same hardware configuration. The UE 10 may include an input unit and an output unit in addition to the hardware configuration shown in Fig. 2. Here, the input unit may include an operation input unit that inputs a user operation (touch panel operation, keyboard operation, etc.) on the UE 10, a microphone that collects the user's voice, etc. The output unit may include a display unit such as a liquid crystal display, a speaker that outputs voice, etc.
Furthermore, the UE 10, the MEC server 22, the management device 23, and the cloud server 40 may each have dedicated hardware for executing their respective functions, or may execute part of all of their functions with hardware and execute the remaining parts with a computer that runs a program. Furthermore, all of the functions may be executed by a computer and a program.

<Functional configuration>
Fig. 3 is an example of a functional configuration of the management device 23. Each function of the management device 23 is, for example, a logical function realized by the hardware of the management device 23 shown in Fig. 2, and can be realized by the CPU 1 executing a program stored in the ROM 2 or the like.
The management device 23 includes a transmitting unit 301 , a receiving unit 302 , a UE information notifying unit 303 , an offload destination determining unit 304 , and an offload control unit 305 .

The transmitting unit 301 and the receiving unit 302 respectively transmit and receive packets via the communication I/F 5 shown in FIG. 2. The UE information notifying unit 303 notifies the cloud server 40 of the identifier of the UE 10 as information (UE information) of the UE 10 accessing the cloud service 100 via the mobile network 20. The offload destination determining unit 304 determines the MEC server 22 to which the UE 10 to be offloaded is to be offloaded based on the offload notification received from the cloud server 40. The offload control unit 305 offloads the UE 10 to be offloaded to the MEC server 22 determined by the offload destination determining unit 304. Offloading control will be described later.

Fig. 4 shows an example of a functional configuration of the MEC server 22. Each function of the MEC server 22 is, for example, a logical function realized by the hardware of the MEC server 22 shown in Fig. 2, and can be realized by the CPU 1 executing a program stored in the ROM 2 or the like.
The MEC server 22 includes a transmitting unit 401 , a receiving unit 402 , a user information acquiring unit 403 , and a content generating unit 404 .

The transmitting unit 401 and the receiving unit 402 respectively transmit and receive packets via the communication I/F 5 shown in FIG. 2. The user information acquiring unit 403 acquires user operation information from the UE 10. The content generating unit 404 generates VR content representing at least a portion of the services of the cloud service 100 based on the user operation information acquired by the user information acquiring unit 403, and delivers (transmits) the VR content to the UE 10 via the communication I/F 5 shown in FIG. 2.

Fig. 5 is an example of a functional configuration of the cloud server 40. Each function of the cloud server 40 is, for example, a logical function realized by the hardware of the cloud server 40 shown in Fig. 2, and can be realized by the CPU 1 executing a program stored in the ROM 2 or the like.
The cloud server 40 includes a transmitting unit 501 , a receiving unit 502 , a user information acquiring unit 503 , a content generating unit 504 , a content database 505 , a UE information acquiring unit 506 , an offload target determining unit 507 , and an offload notifying unit 508 .

The transmitting unit 501 and the receiving unit 502 respectively transmit and receive packets via the communication I/F 5 shown in Fig. 2. The user information acquiring unit 503 acquires user operation information from the UE 10. The content generating unit 504 refers to the content database 505 based on the user operation information acquired by the user information acquiring unit 503, generates VR content representing the cloud service 100, and delivers (transmits) it to the UE 10 via the communication I/F 5 shown in Fig. 2.
The UE information acquisition unit 506 acquires UE information notified by the UE information notification unit 303 of the management device 23. The offload target determination unit 507 determines, among the UEs 10 accessing the cloud service 100, one or more UEs 10 to be offloaded to the MEC server 22. The offload notification unit 508 transmits an offload notification including information indicating the UEs 10 to be offloaded determined by the offload target determination unit 507 to the management device 23.

FIG. 6 shows an example of a virtual game space 110 provided by the cloud service 100 .
FIG. 6 shows an example in which eight users are participating in an online game, and avatars A to H corresponding to the eight users are placed in the virtual game space 110.
The virtual game space 110 is made up of a plurality of areas (fields), and each avatar may be able to move within one area and also between areas.

The virtual game space 110 has, as areas in which avatars can move, for example, a meeting place 111 and multiple battlefields 112 and 113. Avatars participating in an online game can gather at the meeting place 111 and communicate with other avatars using a chat function or the like, or change the equipment of their avatars.
Then, the avatars gathered at the meeting place 111 can move alone or in groups to the battlefields 112 and 113. However, it is not necessary for avatars to be able to move between the battlefields 112 and 113. The battlefields 112 and 113 may each be separate and independent areas.

FIG. 6 shows an example in which Avatar A and Avatar B have moved to a first battlefield (Battlefield #1) 112, and Avatar D has moved to a second battlefield (Battlefield #2) 113.
In this case, only avatars A and B are placed in the first battlefield 112. Avatars C to H are placed in the virtual game space 110, but are placed in areas (the gathering place 111 and the battlefield 113) different from the first battlefield 112 and are therefore invisible to avatars A and B.

The avatars A and B can move automatically within the battlefield 112 and cooperate with each other to fight against enemy characters. Here, the enemy characters may be non-player characters whose actions are automatically controlled. When the enemy characters are defeated, the mission is accomplished and the avatars A and B can return to the meeting place 111.
The number of battlefields is not limited to two, and multiple battlefields may be constructed. The number of gathering places is not limited to one, and multiple gathering places may be constructed. In this case, it is not necessary to allow avatars to move between gathering places.

As described above, a large number of users may log in to the cloud service 100, such as the online game, via various networks. In this case, if a large number of users simultaneously log in to the cloud service 100, the load on the cloud server 40 increases.
Therefore, in this embodiment, the process requested by a specific UE 10 to the cloud server 40 is executed by the MEC server 22 arranged in the mobile network 20, thereby distributing the load (offloading). Specifically, the process is offloaded by transferring a packet that the offload target UE 10 transmits with the cloud server 40 (cloud service 100) as its destination to the MEC server 22 that is the offload destination.

Within the virtual space provided by the cloud service 100, there are a number of independent areas, and in each area, several avatars move and often remain for periods ranging from several minutes to several tens of minutes.
Therefore, in this embodiment, a partial area in the virtual space provided by the cloud service 100 is constructed (reproduced) in the MEC server 22, and an avatar that has requested to move to the partial area is placed in the area constructed in the MEC server 22. Then, the MEC server 22 becomes a service providing server and provides a partial service of the cloud service 100 to a specific UE 10.

Here, the UE 10 to be offloaded is assumed to be a UE that is accessing the cloud service 100 via the mobile network 20 .
1 can access the cloud service 100 via a network other than the mobile network 20. Therefore, when the UEs 10a and 10b access the cloud service 100 via the mobile network 20, the management device 23 detects this and notifies the cloud server 40. At this time, the management device 23 notifies the cloud server 40 of the identifiers of the UEs 10a and 10b.

The cloud server 40 can grasp the UE 10 that is accessing the cloud service 100 via the mobile network 20 by the notification from the management device 23. The cloud server 40 determines whether or not the avatar to be moved to a part of the cloud service 100 is an avatar corresponding to the user that is accessing via the mobile network 20, and when it determines that the access is via the mobile network 20, decides to use the MEC server 22 and transmits an offload notification to the management device 23.
The management device 23 that has received the offload notification transfers packets that are being sent by the UE 10 that is the offload target and is specified by the offload notification, with the cloud service 100 as the destination, to the MEC server 22 that is the offload destination.

In this embodiment, the above-mentioned partial area constructed in the MEC server 22 may be one battlefield.
In this embodiment, when a user requests to move an avatar from a gathering place 111 in the virtual game space 110 of an online game to a battlefield, the above-mentioned battlefield is constructed in the MEC server 22 and the avatar is moved to the battlefield constructed in the MEC server 22 only if the user is accessing the online game via the mobile network 20.

FIG. 7 is a diagram illustrating the offloading control according to the present embodiment.
As shown in this Figure 7, users using UE 10a to 10d are logged in to the cloud service, and avatars A to D corresponding to the users using UE 10a to 10d, respectively, are placed in the virtual game space 110 on the cloud 30.
When avatars A and B corresponding to users of UEs 10a and 10b accessing via mobile network 20 are to be moved to the battlefield, cloud server 40 decides to use MEC server 22 and transmits an offload notification to management device 23. Having received the offload notification, management device 23 constructs a battlefield (field #MEC) 121 in MEC server 22 and places avatars A and B corresponding to users of UEs 10a and 10b in the battlefield 121.
When Avatars A and B finish their stay on the battlefield 121 , they return to the virtual game space 110 on the cloud 30 .

On the other hand, when avatar A corresponding to a user of UE 10a accessing via mobile network 20 and avatar C corresponding to a user of UE 10c accessing via another network, mobile network 50, are moved to the battlefield, the cloud server 40 decides not to use the MEC server 22. In this case, the cloud server 40 builds a battlefield (field #Cloud) 122 on the cloud 30, and places avatars A and C corresponding to users of UE 10a and 10c in the battlefield 122.

Multiple MEC servers 22 may be deployed in the mobile network 20.
When a plurality of MEC servers 22 are arranged in the mobile network 20, the management device 23 may receive an offload notification from the cloud server 40 and determine an offload destination MEC server 22 from among the plurality of MEC servers 22. For example, when the management device 23 receives an offload notification from the cloud server 40, the management device 23 may determine an offload destination MEC server 22 from among the plurality of MEC servers 22 based on information on a connection base station of the UE 10 to be offloaded.

FIG. 8 is a diagram showing an example of the configuration of a mobile network 20 in which a plurality of MEC servers 22a to 22c are arranged.
8, the MEC server 22a is arranged on the route between the UEs 10a, 10b and the MEC server 22b, and the MEC server 22c is arranged on the route between the MEC server 22b and the cloud server 40. In other words, the MEC server 22a is a local MEC arranged at a position closer to the UE 10, the MEC server 22c is a central MEC arranged at a position closer to the center of the mobile network 20, and the MEC server 22b is a middle MEC arranged between the local MEC and the central MEC.

When the number of offload target UEs specified by the offload notification is one, the management device 23 may determine the MEC server that is the shortest distance from the connection base station of the offload target UE as the offload destination MEC server. For example, when the offload target UE is UE 10a whose connection base station is base station 21a in Hokkaido, the management device 23 determines the MEC server 22a that is the shortest distance from the base station 21a among the MEC servers 22a to 22c as the offload destination MEC server.
Here, the distance is not limited to a physical distance, but may be a logical distance based on the number of intermediate nodes present.

In addition, when offloading multiple UEs to the same MEC server, the management device 23 may determine the MEC server that has the smallest sum of distances between the multiple UEs to be offloaded and each of the connected base stations as the offload destination MEC server.
For example, when the offload target UEs are UEs 10a and 10b whose connection base stations are base stations 21a and 21b in Hokkaido, the MEC server 22a, which is a local MEC, is determined as the offload target MEC server. In this case, a battlefield (field #MEC) 121a is constructed in the MEC server 22a, and avatars A and B are placed in the battlefield 121a.

Similarly, when the offload target UEs are UE 10a whose connection base station is the base station 21a in Hokkaido and UE 10e whose connection base station is the base station 21e in Tokyo, the MEC server 22b, which is the middle MEC, is determined as the MEC server of the offload destination. In this case, a battlefield (field #MEC) 121b is constructed in the MEC server 22b, and avatars A and E are arranged in the battlefield 121b.
In addition, when the offload target UEs are UE 10a whose connection base station is the base station 21a in Hokkaido and UE 10f whose connection base station is the base station 21f in Okinawa, the MEC server 22c, which is the central MEC, is determined as the offload destination MEC server. In this case, a battlefield (field #MEC) 121c is constructed in the MEC server 22c, and avatars A and F are arranged in the battlefield 121c.

Furthermore, when the mobile network 20 is configured as a ring network, the management device 23 may determine the offload destination MEC server in consideration of the ring network to which the connected base station of the UE to be offloaded belongs.
9 shows an example of the configuration of a ring network. As shown in this figure, a mobile network 20 is made up of a plurality of ring networks, and includes a plurality of base stations 21 and a plurality of accommodation stations 24 to 26.

Accommodation station 24 is an edge data center (hereinafter also referred to as a "GC (Group unit Center)"), accommodation station 25 is a regional data center (Regional Data Center: RDC), and accommodation station 26 is a central data center (Central Data Center: CDC). Note that the black squares in FIG. 9 indicate GC 24 or RDC 25. A backhaul network (Mobile Backhaul: MBH) is formed between GC 24 connected to base station 21 and CDC 26.

The GC 24 is installed near the base station 21 and can be connected to each of the multiple base stations 21. The RDC 25 is connected to multiple GCs 24 arranged in a target area. The GC 24 or the RDC 25 performs the functions of a Distributed Unit (DU) and a Central Unit (CU), which are part of the functions of the RAN.
The CDC 26 is a large-scale data center connected to a plurality of GCs 24 and a plurality of RDCs 25. This CDC 26 performs the function of a 5G core network.
As shown in FIG. 9, a plurality of base stations 21 and a CDC 26 are connected in a ring shape.

A GC ring is one of the ring networks constituting the mobile network 20, and aggregates GCs 24 that are directly connected to the base station 21. An M ring is a higher-level ring network than the GC ring, and aggregates GC rings in specific area units (for example, prefecture units). An L ring is a higher-level ring network than the M ring, and aggregates multiple M rings and connects them to a CDC 26. There may also be GCs 24 that belong to an M ring or an L ring without going through a GC ring.
The MEC server 22 can be arranged for each ring, for example. In this case, the MEC server 22 can be arranged in an accommodation station such as a GC 24, an RDC 25, or a CDC 26. In the example shown in Fig. 9, the MEC server 22a, which is a local MEC, is arranged in the GC 24, the MEC server 22b, which is a middle MEC, is arranged in the RDC 25, and the MEC server 22c, which is a central MEC, is arranged in the CDC 26.
The number of MEC servers 22 arranged in one ring is not particularly limited. A plurality of MEC servers 22 may be arranged in one ring, and a ring may exist in which no MEC server 22 is arranged.

In this way, when the mobile network 20 is configured as a ring network, the management device 23 may determine the offload destination MEC server 22 from among the MEC servers 22 belonging to the ring network to which the connected base station of the UE 10 to be offloaded belongs.
For example, when the UEs to be offloaded are UEs 10a and 10b whose connected base station is a base station 21 belonging to the same GC ring R1, the MEC server 22a belonging to the GC ring R1 may be determined as the offload destination MEC server.
In addition, if the UEs to be offloaded are UEs 10a and 10e that belong to different GC rings and different M rings, respectively, and have a base station 21 belonging to the same L ring R2 as their connected base station, the MEC server 22b belonging to L ring R2 may be determined as the MEC server to be offloaded to.
Furthermore, when the UE to be offloaded is UE 10a, 10f, whose connected base stations are base stations 21 belonging to different L rings R2, R3, respectively, the MEC server 22c accommodated in CDC 26 may be determined as the MEC server to be offloaded.

Furthermore, the mobile network 20 may be configured to be capable of controlling end-to-end network slicing across the RAN, backhaul network, and core network. Network slicing is a network architecture in which a network is virtually divided (sliced) according to the intended use, operated as slices, and services according to the intended use are provided.

FIG. 10 is a conceptual diagram of network slicing control.
FIG. 10 shows an example in which a UE 10a, a paying user (guaranteed bandwidth user), normally accesses the cloud server 40 using a high-quality slice 27a, and a UE 10b, a non-paying user, accesses the cloud server 40 using a standard slice 27b.
When the UE 10a of a paying user and the UE 10b of a non-paying user are offloaded to the same MEC server 22a as offload target UEs, the slice levels set for the UEs 10a and 10b may be set to the highest quality slice level among them, that is, the high quality slice level of the UE 10a. In this case, both the UEs 10a and 10b use the high quality slice 27a.

UE 10a, 10b may be capable of setting multiple slices with different slice levels, and UE 10b may use high-quality slice 27a only when using MEC server 22a, and may use standard slice 27b when using other applications, etc. Also, high-quality slices may be used to stabilize communication quality only for services that require real-time performance.

Next, a packet control method in this embodiment will be described.
As described above, when the management device 23 receives an offload notification from the cloud server 40, the management device 23 transfers to the MEC server 22 packets transmitted from the UE 10 that is the offload target specified by the offload notification to the cloud server 40.
In this embodiment, a UPF (User Plane Function) connected to the MEC server 22 is controlled so that the MEC server 22 can receive packets transmitted by the UE 10 to be offloaded without the cloud server 40 receiving the packets.

FIG. 11 is a schematic diagram of a portion of a 5G network.
11 includes a UE 201, a RAN 202, UPFs 203 to 206, and MECs 207 to 209. The UPF 203 is connected to the MEC 207, the UPF 204 is connected to the MEC 208, the UPF 205 is connected to the MEC 209, and the UPF 206 is connected to a DN (Data Network) 210.
UE 201 corresponds to UEs 10a and 10b in Fig. 1. RAN 202 includes base station 21 in Fig. 1. MECs 207 to 209 correspond to MEC servers 22a to 22c in Fig. 8. DN 210 is a network such as the Internet or a data center, and corresponds to cloud 30 in Fig. 1.

The management device 23 can change the route of a packet from the UE 201 to the DN 210 to any of the MECs 207 to 209 by controlling the UPFs 203 to 205 connected to the MECs 207 to 209 (intercept control).
In other words, if the management device 23 does not instruct the UPFs 203 to 205 to offload, a packet sent from the UE 201 and addressed to the DN 210 is sent to the DN 210 via the route indicated by the arrow 221 .

Then, when the management device 23 instructs the UPF 204 to offload, for example, the packet sent from the UE 201 and addressed to the DN 210 is changed in route as shown by the arrow 222 and sent to the MEC 208 .
In this case, when UPF 204 detects a packet whose source address is the IP address of UE 10 to be offloaded specified by the management device 23 and whose destination address is the IP address of cloud server 40, it forwards the packet to MEC 208.

In addition, since the UPFs 203 to 206 are controlled by an SMF (Session Management Function), the management device 23 may change the route of a packet to the DN 210 to any one of the MECs 207 to 209 by controlling the SMF.
In addition, the UPF is a function of the core network and is not usually provided in the RAN. Therefore, when the offload destination MEC server is arranged in the RAN (base station), the UPF may be provided separately in the RAN.

<Processing flow>
A specific description will be given below of a process flow in the communication system 1 of the present embodiment. Here, a process flow in a case where the UE 10a accesses the cloud service 100 via the mobile network 20 and receives the provision of the service will be described.
First, as a preparation, a virtual environment is constructed in the MEC server 22 so that a partial area (here, a battlefield) of the cloud service 100 can be reproduced. At this time, if the partial area to be constructed in the MEC server 22 is determined in advance, the partial area may be constructed in advance.

When the UE 10a connects to the mobile network 20, the UE 10a transmits a registration request message to the mobile network 20. This registration request message may be, for example, a "REGISTRATION REQUEST" defined in 3GPP.
The registration request message transmitted from the UE 10a is transmitted to the core network via the RAN of the mobile network 20, and connection with the UE 10a is permitted after some processing in the core network. Then, a registration approval message is transmitted from the core network to the UE 10a, and the UE 10a receives this. This registration approval message may be, for example, "REGISTRATION ACCEPT" as defined by 3GPP.

At this time, the orchestrator of the core network registers information about the connected base station of the UE 10a in association with the identifier of the UE 10a. Here, the identifier of the UE 10a may include an IP address, SIM information (SIM ID), etc. Also, the information about the connected base station may include, for example, a base station ID (Cell ID: CID), an affiliated ring ID, etc.
In this manner, a PDU (Packet Data Unit) session for the UE 10a to connect to an external service is established. That is, a PDU session is established between the UE 201 shown in FIG. 11 and the UPF 206 connected to the DN 210, and the UE 201 is in a state in which it can communicate with a server in the DN 210 via the path 221.

FIG. 12 is a communication sequence diagram after the UE 10 a connects to the mobile network 20 .
First, the UE 10a transmits a login request to connect to the cloud service 100. Specifically, in step S1, a login request packet is transmitted from the UE 10a to the cloud server 40. This packet is transmitted to the cloud server 40 via the mobile network 20.

At this time, the management device 23 detects that a packet addressed to the cloud server 40 has been sent from the UE 10a. Specifically, the UPF detects the packet addressed to the cloud server 40 based on the IP address, and while sending the packet to the cloud server 40, notifies the management device 23 that the packet addressed to the cloud server 40 has been received. Note that the IP address of the cloud server 40 is assumed to be registered in advance.

In step S2, upon receiving the notification from the UPF, the management device 23 notifies the cloud server 40 that a packet (login request) from the UE 10a has been sent via the mobile network 20. At this time, the management device 23 transmits the identifier of the UE 10a as information about the UE 10a. The identifier of the UE 10a may include at least one of the IP address and SIM ID of the UE 10a. The management device 23 may also transmit attribute information of the UE 10a together with the identifier of the UE 10a. The attribute information of the UE 10a may include at least one of the CID, the ring ID to which the UE 10a belongs, and the configured slice level (high, standard, etc.) of the UE 10a.

In step S3, the cloud server 40 performs a login process for the UE 10a to the cloud service 100. The cloud server 40 also registers that the UE 10a is accessing the cloud service 100 via the mobile network 20.
When the login process is completed, the cloud server 40 starts providing the service to the UE 10a. Specifically, the cloud server 40 generates content in which an avatar corresponding to the user of the UE 10a is arranged in the virtual game space 110 (step S4), and distributes the content to the UE 10a. The UE 10a outputs (displays) the content distributed from the cloud server 40 (step S5). Then, the UE 10a transmits information (user operation information) indicating an operation input to the UE 10a by a user who viewed the content to the cloud server 40. Then, the cloud server 40 generates content based on the user operation information (step S4), and distributes the content to the UE 10a. In this manner, the cloud service 100 is provided from the cloud server 40 to the UE 10a.

Note that while the cloud server 40 is providing services to the UE 10a, the cloud server 40 may transmit information such as an avatar corresponding to the user of the UE 10a to the management device 23 as necessary. For example, when the avatar corresponding to the user of the UE 10a is changed (updated), the cloud server 40 may transmit the latest avatar information to the management device 23. The management device 23 can store the information received from the cloud server 40 in a specified location and use the information when constructing a battlefield in the MEC server 22.

In the virtual game space 110 on the cloud 30, when the user of the UE 10a requests to move to the battlefield, a field movement request is transmitted from the UE 10a to the cloud server 40 in step S6.
When the cloud server 40 receives the field movement request from the UE 10a, in step S7, based on the identifier of the UE 10a, the cloud server 40 refers to the UE information registered in step S3 and determines whether or not the UE 10a is accessing via the mobile network 20. Then, when the cloud server 40 confirms that the UE 10a is accessing via the mobile network 20, in step S7, the cloud server 40 decides to use the MEC server 22 and determines the UE 10a as a UE to be offloaded.

Note that if the connection between UE 10a and mobile network 20 is cut off, for example by UE 10a switching to a Wi-Fi connection, this can be recognized by the mobile network 20. In this case, management device 23 may notify cloud server 40 of the identifier of UE 10a and that UE 10a is no longer accessing via mobile network 20. This allows cloud server 40 to cancel the registration performed in step S3. As a result, it is possible to avoid UE 10a being mistakenly determined to be an offload target UE.

In step S8, the cloud server 40 transmits an offload notification to the management device 23. The offload notification may include an identifier of the UE 10a as UE information of the offload target. The offload notification may also include information about the battlefield to be constructed, avatar information corresponding to the user of the UE 10 of the offload target, and the like.
In step S9, the management device 23 that has received the offload notification determines the MEC server 22 to be the offload destination. At this time, as described above, the management device 23 may determine the MEC server 22 to be the offload destination from among the multiple MEC servers 22 arranged in the mobile network 20 based on information about the connection base station of the UE 10a. Note that information about the connection base station of the UE 10a is registered in the orchestrator in association with the identifier of the UE 10a when the UE 10a connects to the mobile network 20. Therefore, the management device 23 can check information about the connection base station of the UE 10a based on the identifier of the UE 10a specified by the offload notification, and appropriately determine the MEC server 22 to be the offload destination.

When the management device 23 determines the offload destination MEC server 22, in step S10, it transmits a field construction instruction to the offload destination MEC server 22. This field construction instruction may include information on the battlefield to be constructed, avatar information to be placed in the constructed battlefield, etc. At this time, the management device 23 also transmits a packet forwarding instruction (not shown) to the UPF connected to the offload destination MEC server 22. This packet forwarding instruction is an instruction to forward a packet transmitted from the UE 10a to the cloud server 40 to the MEC server 22.
In step S11, the MEC server 22 that has received the field construction instruction constructs the specified battlefield and places an avatar corresponding to the user of the UE 10a. Then, the MEC server 22 starts providing the service to the UE 10a.

Specifically, the MEC server 22 generates content in which an avatar corresponding to the user of the UE 10a is placed on a battlefield (step S12), and distributes the content to the UE 10a. The UE 10a outputs (displays) the content distributed from the MEC server 22 (step S13). Then, the UE 10a transmits user operation information to the cloud server 40. The packet addressed to the cloud server 40 is routed by the UPF connected to the MEC server 22 and transferred to the MEC server 22 instead of the cloud server 40. Therefore, the MEC server 22, not the cloud server 40, generates content based on the user operation information (step S12), and distributes the content to the UE 10a.
In this manner, some of the services of the cloud service 100 are provided to the UE 10a from the MEC server 22. At this time, the user of the UE 10a can enjoy the services as if he or she were logged in to the cloud service 100.

In the above example, the offload destination MEC server 22 is determined based on information about the base station connected to the UE 10a. However, the offload destination MEC server 22 may be determined based on the processing content of the offload target. For example, if the processing for constructing the battlefield to which the UE 10a is about to move is a processing that imposes a relatively large processing load on the server, an MEC server 22 having specifications that match the processing content may be selected from among the multiple MEC servers 22 and determined as the offload destination MEC server 22. Also, among the multiple MEC servers 22, an MEC server 22 with a small current processing load (processing volume, resource usage rate) may be determined as the offload destination MEC server 22. Furthermore, if the battlefields that can be reproduced are determined for each MEC server 22, the offload destination MEC server 22 may be determined from among the MEC servers 22 that can reproduce the battlefield to which the UE 10a is about to move.

Furthermore, in the above example, the case where there is one UE to be offloaded has been described, but there may be multiple UEs to be offloaded. In this case, the cloud server 40 may determine whether all of the multiple UEs 10 sending the field movement request are accessing via the mobile network 20, and if all of the UEs 10 are accessing via the mobile network 20, may decide to offload these UEs 10 to the same MEC server 22.

Furthermore, when the cloud server 40 receives field movement requests from a plurality of UEs 10, the cloud server 40 may determine a group including one or more UEs 10 to be offloaded to the same MEC server 22 from among the plurality of UEs 10. In this case, the cloud server 40 may preferentially group the UEs 10 that are accessing the cloud server 40 via the mobile network 20 based on the identifier of each UE 10.
Furthermore, the cloud server 40 may group the UEs to be offloaded based on attribute information of the UEs 10. For example, the cloud server 40 may preferentially group the UEs 10 that have the same or similar connected base station or ring ID. The cloud server 40 may also preferentially group the UEs 10 that have the same set slice level.

Returning to FIG. 12, when the user ends the operation on the battlefield, for example by defeating an enemy character in the battlefield and completing a mission, the MEC server 22 detects the end of the operation in step S14, and notifies the cloud server 40 of the result in step S15. The result notified to the cloud server 40 may include, for example, game play history information on the battlefield, the latest avatar information, etc.

When cloud server 40 receives the result notification from MEC server 22, it updates the content database, etc. based on the received information. Then, cloud server 40 transmits an offload end instruction to management device 23 in step S16.
In step S17, the management device 23 that has received the offload end instruction instructs the MEC server 22 to perform end processing, and in step S18, the MEC server 22 executes the end processing. This end processing may include, for example, a process of deleting a battlefield that has been constructed for offloading.
At this time, the management device 23 also transmits a packet forwarding end instruction (not shown) to the UPF connected to the MEC server 22 that is to be terminated. This packet forwarding end instruction is an instruction to restore the packet transmitted from the UE 10a to the cloud server 40 so that the packet is transmitted to the cloud server 40.

As a result, the transfer of packets by the UPF to the MEC server 22 is terminated, and the provision of services is switched to the cloud server 40 being provided to the UE 10a.
That is, similar to steps S4 and S5 described above, the cloud server 40 generates content in which an avatar corresponding to the user of the UE 10a is placed in the virtual game space 110 (step S19), and distributes the content to the UE 10a. The UE 10a outputs (displays) the content distributed from the cloud server 40 (step S20). The UE 10a also transmits user operation information to the cloud server 40. The cloud server 40 then generates content based on the user operation information (step S19), and distributes the content to the UE 10a.

In the above example, the mobile network 20 controls the transfer of packets addressed to the cloud server 40 from the offload target UE 10a to the MEC server 22. However, the UE 10a may control packet transmission. In this case, when the management device 23 receives an offload notification from the cloud server 40, it notifies the UE 10a to transmit packets addressed to the cloud server 40 to the MEC server 22. The UE 10a that receives the notification uses pre-installed software (application) to change the packets addressed to the cloud server 40 to packets addressed to the MEC server 22 and transmits the changed packets. For example, the software (application) may perform a process of switching the settings of the DNS (Domain Name System). As a result, packets addressed to the cloud server 40 from the UE 10a are transmitted to the MEC server 22 as the destination.

As described above, in the communication system 1 of this embodiment, when the management device 23 in the mobile network 20 detects a packet addressed to the cloud service 100, it notifies the cloud service 100 of the identifier of the UE 10 that sent the packet. Furthermore, when the management device 23 receives an offload notification from the cloud service 100 to offload at least a portion of the services of the cloud service 100 to the UE 10, the management device 23 forwards the packet addressed to the cloud service 100, with the UE 10 to be offloaded specified by the offload notification as the source, to the MEC server 22 arranged in the mobile network 20. As a result, at least a portion of the services of the cloud service 100 are provided to the UE 10 using the MEC server 22.

In this way, in the communication system 1 of the present embodiment, some services can be provided by using the MEC server 22, and therefore the processing load of the cloud server 40 can be reduced. Furthermore, by using the MEC server 22 arranged in the mobile network 20, the distance between the UE 10 and the service providing server can be shortened and delays can be reduced, compared to the case where the cloud server 40 is used as the service providing server.
Furthermore, in the communication system 1 of the present embodiment, whether or not to offload the UE 10 to the MEC server 22 is determined depending on an access route from the UE 10 to the cloud service 100. In this communication system 1, the UE 10 accessing the cloud service 100 via the mobile network 20 is determined as the UE 10 to be offloaded, and is offloaded to the MEC server 22. Therefore, it is possible to appropriately realize the provision of some services using the MEC server 22 arranged in the mobile network 20.

Furthermore, users do not always use the same network, and the cloud server 40 cannot determine, from pre-registration information or terminal information, via which network each UE 10 is accessing the cloud service 100. In this embodiment, the management device 23 of the mobile network 20 notifies the cloud server 40 that the UE 10 is accessing the cloud server 40 via the mobile network 20. At this time, the management device 23 can transmit at least one of an IP address and SIM information as an identifier of the UE 10 accessing the cloud server 40 via the mobile network 20.
This allows the cloud server 40 to properly grasp the UE 10 that is accessing via the mobile network 20. Therefore, the cloud server 40 can properly transmit to the management device 23 an offload notification for offloading only the UE 10 that is accessing via the mobile network 20. In this way, the cloud server 40 can cause the management device 23 to properly perform offloading control.

Packet forwarding to the offload destination MEC server 22 can be achieved by controlling the UPF connected to the MEC server 22. In this way, the UPF is controlled to forward packets destined for the cloud service 100 to the MEC server 22, so the UE 10 does not need to change the packet transmission process before and after offloading.

Here, the cloud service 100 may be a service that provides content in which an avatar corresponding to a user of the UE 10 is placed in a virtual space. For example, the cloud service 100 may be an online game that provides a virtual game space 110.
In this case, the MEC server 22 may construct a battlefield, which is a partial area of the virtual game space 110, and place avatars corresponding to users of the UEs 10 to be offloaded, in the constructed battlefield. The battlefield of an online game is an area in which one or several avatars act independently of other avatars, and therefore can be easily and appropriately assigned to the MEC server 22.

In addition, when the management device 23 receives an offload notification from the cloud server 40, it can determine the MEC server 22 to which the offload is to be performed based on information about the connected base station of the UE 10 to be offloaded, which is specified by the offload notification.
For example, when there is one UE 10 to be offloaded, the management device 23 may determine, as the offload destination edge server, the MEC server 22 that is the closest to the connected base station of the UE 10 to be offloaded among the multiple MEC servers 22. This allows the service to be provided from the MEC server 22 that is closest to the UE 10, thereby significantly improving the response time.

Furthermore, when multiple UEs 10 are offloaded to the same MEC server 22, the management device 23 may determine, among the multiple MEC servers 22, the MEC server 22 having the smallest sum of distances to the respective connected base stations of the multiple UEs 10 to be offloaded as the offload destination MEC server 22. This makes it possible to appropriately improve the response time of each UE 10.
The distance may be a physical distance or a logical distance.

Furthermore, the management device 23 may determine the offload destination MEC server 22 from among the MEC servers 22 belonging to the ring network to which the connected base station of the offload target UE 10 belongs. In this manner, the offload destination MEC server 22 may be determined depending on which ring network the base station to which the offload target UE 10 is connected belongs to.
Even in this case, it is possible to appropriately improve the response time in each UE 10. Moreover, it is possible to relatively easily determine the offload destination MEC server 22 based on the ring ID of the connected base station of the offload target UE 10.

In addition, when the management device 23 offloads multiple communication devices to the same edge server, if the connection base stations of the communication devices to be offloaded belong to different ring networks, the management device 23 may determine the offload destination edge server from among the edge servers belonging to the highest ring network among the ring networks to which the connection base stations of the communication devices to be offloaded belong. Alternatively, the offload destination edge server may be determined from among the edge servers belonging to an even higher ring network. For example, when offloading a first communication device belonging to a first GC ring and a second communication device belonging to a second GC ring, the offload destination edge server may be determined from among the edge servers belonging to the M ring that aggregates the first GC ring and the second GC ring, or the L ring that aggregates the M ring. This makes it possible to equalize the response times of the multiple communication devices to be offloaded.

In addition, when the management device 23 offloads multiple UEs 10 to the same MEC server 22, the management device 23 may set the slice level of each UE 10 to be offloaded to the slice level with the highest quality among them. For example, when a paying user who uses a high-quality slice and a non-paying user who uses a standard slice use the same MEC server 22, the high-quality slice is used. This allows the paying user to continue receiving the service without feeling a decrease in communication quality.

Furthermore, in this embodiment, the cloud server 40 may determine the UE 10 to be offloaded based on the identifier of the UE 10 that is accessing via the mobile network 20 and that is notified by the management device 23 .
In the above online game, when a movement to the battlefield is requested from a plurality of UEs 10, the cloud server 40 may preferentially group the UEs 10 accessing the battlefield via the mobile network 20 based on the identifiers of the UEs 10, and may determine to offload the UEs 10 to the same MEC server 22. This can actively reduce the processing load on the cloud server 40.

Furthermore, the cloud server 40 may determine a plurality of UEs 10 to be offloaded to the same MEC server 22 based on attribute information of the UEs 10 accessing the cloud server 40 via the mobile network 20, which is notified from the management device 23. Here, the attribute information may include at least one of the CID, the belonging ring ID, and the configured slice level.
For example, in the above online game, when a movement to the battlefield is requested from a plurality of UEs 10, the cloud server 40 may determine, based on the CID or the ring ID of each UE 10, to preferentially group UEs 10 that connect to the same or nearby base station and offload them to the same MEC server 22. This can actively reduce the processing load of the cloud server 40 while improving the response time of each UE 10.

Furthermore, in the above online game, when a movement to the battlefield is requested from a plurality of UEs 10, the cloud server 40 may determine, based on the slice level setting of each UE 10, to preferentially group UEs 10 having the same slice level setting and offload the UEs 10 to the same MEC server 22. This makes it possible to actively reduce the processing load of the cloud server 40 while maintaining the communication quality of each UE 10 as much as possible.
Furthermore, in the case of the above online game, during a battle in the battlefield, avatar A may request help from other avatars outside the battlefield. In other words, avatar B, which receives a help request later, may move to the battlefield of the MEC server 22 where avatar A is located. In this case, the cloud server 40 may determine, among the UEs 10 accessing via the mobile network 20, a UE 10 whose connected base station is the same as or close to the connected base station of the UE 10 corresponding to avatar A as a UE 10 to be offloaded later, or may determine, as a UE 10 whose set slice level is equivalent to the set slice level of the UE 10 corresponding to avatar A, as a UE 10 to be offloaded later.

As described above, in this embodiment, some of the services of the cloud service 100 are offloaded to the MEC server, so the processing load on the cloud server 40 that provides the cloud service 100 can be reduced. Also, in this embodiment, the UE 10 that is accessing the cloud service 100 via the mobile network 20 is offloaded to the MEC server 22, so the operability for users who use the mobile network 20 can be improved compared to users who use other networks.

In the above embodiment, the cloud service 100 is a service that provides a virtual game space for an online game, but the present invention is not limited to the above. The cloud service 100 may be any service that allows a large number of users to log in via various networks including at least the mobile network 20.
For example, the cloud service 100 may be an online competitive game such as Go or Shogi. In this case, a competition area may be constructed in the MEC server 22, and the UEs 10 of the two players competing may be offloaded.

Note that although specific embodiments have been described above, these embodiments are merely examples and are not intended to limit the scope of the present disclosure. The devices and methods described herein may be embodied in forms other than those described above. Furthermore, omissions, substitutions and modifications may be made to the above-described embodiments as appropriate without departing from the scope of the present disclosure. Forms in which such omissions, substitutions and modifications have been made are included within the scope of those described in the claims and their equivalents, and belong to the technical scope of the present disclosure.

(Embodiments of the present disclosure)
The present disclosure includes the following embodiments.
[1] A communication system comprising one or more processors, wherein at least one of the one or more processors executes the following steps when a packet addressed to a specified destination outside a communication network that relays packets is detected: a notification process for notifying the specified destination of an identifier of a communication device that transmitted the packet; a reception process for receiving from the specified destination an offload notification for offloading at least a portion of services for the specified destination to the communication device; and a forwarding process for forwarding a packet addressed to the specified destination, from the communication device that is an offload target specified by the offload notification as a source, to an edge server disposed in the communication network that provides at least a portion of services for the specified destination.

[2] The communication system described in [1], further characterized in that at least one of the one or more processors executes a first determination process for determining an edge server to which the offload is to be performed from among the edge servers arranged in the communication network, based on information regarding a base station connected to the communication device to be offloaded.

[3] The communication system described in [2], characterized in that, in the first determination process, when there is one communication device to be offloaded, the edge server among the multiple edge servers that is the shortest distance from the connected base station of the communication device to be offloaded is determined as the edge server to be the offload destination.

[4] The communication system described in [2] or [3], characterized in that, in the first determination process, when multiple communication devices are to be offloaded to the same edge server, the edge server among the multiple edge servers that has the smallest sum of distances to each of the connected base stations of the multiple communication devices to be offloaded is determined as the offload destination edge server.

[5] The communication system according to any one of [2] to [4], characterized in that the first determination process determines the edge server to which the offload is to be performed from among edge servers that belong to a ring network to which the connected base station of the communication device to be offloaded belongs.

[6] The communication system according to any one of [2] to [5], characterized in that, in the first determination process, when multiple communication devices are offloaded to the same edge server, if the connected base stations of the communication devices to be offloaded belong to different ring networks, the offload destination edge server is determined from among edge servers belonging to a higher-level ring network that aggregates the ring networks to which the connected base stations belong, or from among edge servers belonging to an even higher-level ring network that aggregates the higher-level ring networks.

[7] The communication system according to any one of [1] to [6], further characterized in that when multiple communication devices are offloaded to the same edge server by at least one of the one or more processors, a slice control process is executed to adjust the setting slice level of each communication device to be offloaded to the slice level with the highest quality among the multiple communication devices.

[8] The communication system described in any one of [1] to [7], characterized in that the forwarding process controls a UPF connected to the edge server that is the offload destination, and forwards packets addressed to the specified destination to the edge server.

[9] The communication system according to any one of [1] to [8], further comprising a second determination process that determines the communication device to be offloaded based on the identifier of the communication device notified to the specified destination, by at least one of the one or more processors.

[10] The communication system described in [9], characterized in that the notification process further notifies the specified destination of attribute information of the communication device, and the second determination process determines, based on the attribute information of the communication device, multiple communication devices to be offloaded to the same edge server as the communication devices to be offloaded.

[11] The communication system described in [10], wherein the notification process notifies at least one of the connected base station ID of the communication device, the ring network ID to which the connected base station belongs, and the set slice level as attribute information of the communication device.

[12] The communication system described in any one of [1] to [11], characterized in that the specified destination is a service that provides content in which an avatar corresponding to a user of the communication device is placed in a virtual space, and a placement process is further executed by at least one of the one or more processors to construct an area of at least a portion of the virtual space in the edge server and place an avatar corresponding to the user of the communication device to be offloaded in that area.

[13] The communication system according to any one of [1] to [12], characterized in that the notification process transmits at least one of the IP address and SIM information of the communication device as an identifier of the communication device.

[14] A communication control method characterized by including, in a communication network that relays packets, when a packet addressed to a predetermined destination outside the communication network is detected, notifying the predetermined destination of an identifier of a communication device that transmitted the packet, receiving an offload notification from the predetermined destination for offloading at least a portion of the services of the predetermined destination to the communication device, and forwarding the packet addressed to the predetermined destination, with the communication device designated by the offload notification as the offload target as the source of the packet, to an edge server disposed in the communication network that provides at least a portion of the services of the predetermined destination.

10a to 10f: UE (communication device), 20: mobile network (communication network), 21: base station, 22: MEC server (edge server), 23: management device, 30: cloud, 40: cloud server, 100: cloud service

Claims (14)

1. One or more processors;
   by at least one of the one or more processors,
   a notification process for notifying a predetermined destination of an identifier of a communication device that has transmitted a packet when a packet addressed to a predetermined destination outside the communication network is detected in a communication network that relays packets;
   a receiving process of receiving an offload notification from the predetermined destination to cause the communication device to offload at least a part of a service of the predetermined destination;
   a forwarding process of forwarding a packet having the communication device as an offload target specified by the offload notification as a source and the predetermined destination as a destination to an edge server disposed in the communication network and providing at least a part of a service of the predetermined destination;
   is executed,
   A communication system comprising:
2. by at least one of the one or more processors,
   The communication system according to claim 1, further comprising a first determination process for determining an edge server to be offloaded from among the edge servers arranged in the communication network, based on information regarding a base station connected to the communication device to be offloaded.
3. The first determination process includes:
   The communication system according to claim 2, characterized in that, when there is one communication device to be offloaded, the edge server among the plurality of edge servers that is the shortest distance from the connected base station of the communication device to be offloaded is determined as the edge server to be the offload destination.
4. The first determination process includes:
   The communication system according to claim 2, characterized in that, when multiple communication devices are offloaded to the same edge server, the edge server among the multiple edge servers that has the smallest sum of distances to each connected base station of the multiple communication devices to be offloaded is determined as the offload destination edge server.
5. The first determination process includes:
   The communication system according to claim 2 , wherein the offload destination edge server is determined from among edge servers that belong to a ring network to which a connected base station of the communication device to be offloaded belongs.
6. The first determination process includes:
   The communication system according to claim 2, characterized in that when offloading multiple communication devices to the same edge server, if the connected base stations of the communication devices to be offloaded belong to different ring networks, the offload destination edge server is determined from among edge servers belonging to an upper ring network that aggregates each ring network to which the connected base stations belong, or an even higher level ring network that aggregates the upper ring networks.
7. by at least one of the one or more processors,
   The communication system described in claim 1, characterized in that when multiple communication devices are offloaded to the same edge server, a slice control process is further executed to adjust the setting slice level of each communication device to be offloaded to the slice level with the highest quality among them.
8. The transfer process includes:
   The communication system according to claim 1 , further comprising: controlling a UPF connected to the edge server as an offload destination, and transferring a packet addressed to the predetermined destination to the edge server.
9. by at least one of the one or more processors,
   The communication system according to claim 1 , further comprising: a second determination process for determining the communication device to be offloaded based on an identifier of the communication device notified to the predetermined destination.
10. The notification process further includes notifying the predetermined destination of attribute information of the communication device.
    The second determination process includes:
    The communication system according to claim 9 , further comprising: determining, as the offload target communication devices, a plurality of the communication devices to be offloaded to the same edge server, based on attribute information of the communication devices.
11. The communication system according to claim 10, characterized in that the notification process notifies at least one of a connected base station ID of the communication device, a ring network ID to which the connected base station belongs, and a set slice level as attribute information of the communication device.
12. the predetermined destination is a service that provides content in which an avatar corresponding to a user of the communication device is placed in a virtual space,
    by at least one of the one or more processors,
    The communication system according to claim 1, further comprising a placement process for constructing at least a portion of the virtual space in the edge server and placing an avatar corresponding to a user of the communication device to be offloaded in the portion of the virtual space.
13. The communication system according to claim 1 , wherein the notification process transmits at least one of an IP address and SIM information of the communication device as an identifier of the communication device.
14. In a communication network that relays packets, when a packet addressed to a predetermined destination outside the communication network is detected, an identifier of the communication device that transmitted the packet is notified to the predetermined destination;
    receiving an offload notification from the predetermined destination to cause the communication device to offload at least a part of a service of the predetermined destination;
    a communication control method comprising: forwarding a packet having the communication device to be offloaded specified by the offload notification as a source and destined for the specified destination to an edge server disposed in the communication network, the edge server providing at least a portion of a service to the specified destination.
    
    
    
    
    

Images