CN117336808A - Re-placement method for guaranteeing MEC service continuity - Google Patents

Abstract

The invention relates to a replay method for guaranteeing MEC service continuity, which comprises the following steps: step S1, mobile terminal MN is on line, registration and session registration flow are completed, session is bound with terminal identity; step S2, the mobile terminal MN is about to switch across gateways, a network layer and an application layer notification cooperative mechanism are triggered, and the application layer perceives the network layer state change; step S3, the mobile terminal MN performs cross-gateway switching, and cooperatively performs network layer switching management and application layer edge application migration flow. The scheme meets the requirements of different scenes and continuity of the service, simplifies the replay flow and network deployment of the MEC service, and trades lower service delay with lower service migration overhead.

Description

A relocation method to ensure MEC business continuity

Technical field

The present invention relates to a method, specifically to a relocation method for ensuring MEC business continuity, and belongs to the technical field of MEC business.

Background technique

In recent years, with the development of cloud-network integration and computing-network integration, network connections have developed from tangible physical connections to intangible virtual connections such as content, services, and computing power, making mobile scenarios and subjects more ubiquitous, including terminals and servers. , ubiquitous mobility of networks, etc.; secondly, the further flattening of networks and the application of high-density networking on the wireless side have made horizontal handovers more frequent. This change in paradigm has introduced a new scenario of multi-network convergence and overlapping coverage areas. In this scenario, users adopt multiple access modes, and network nodes and user terminals move at high speed. There is an urgent need to enhance the ability of mobile nodes to move between multiple access networks. Seamless switching capabilities to ensure business continuity.

Due to the existence of fixed mobility anchors in the existing mobile network architecture, the existing MEC (Multi-access Edge Computing) relocation method cannot well guarantee business continuity in mobile states. A typical scenario: high-speed vehicles frequently switch across gateways, making it difficult to ensure business continuity. At the same time, ultra-low latency and ultra-high reliability cannot be met. MEC coverage is limited, and frequent cross-gateway switching of vehicles makes it difficult to support business continuity. ETSI (European Telecommunications Standards Institute, European Telecommunications Standards Institute) and 3GPP (3rd Generation Partnership Project, 3rd Generation Partnership Project) have proposed some MEC business relocation solutions. ETSI's MEC service relocation involves MA (Mobility Anchor) in the network layer and ME (Mobile Edge) APP (Application) in the application layer. There are two solutions. Option 1: MA switching, APP migration; Option 2: MA switching, APP not migrated. 3GPP proposed three modes of SSC (Session and Service Continuity).

In ETSI solution 1, the MN (Mobile Node, that is, the mobile terminal) leaves the service range of the home MA and moves to the service range of the T-MA (Target MA, that is, the destination mobility anchor). First, the MN performs signal detection and sends a routing request to the MA with the strongest signal through RAN (RadioAccess Network), and the MA with the strongest signal is set as T-MA. Then, after T-MA receives the routing request, T-MA communicates with AME (Access Manager Network Element, access and mobility management network element) and SME (Session Manager Network Element, session management network element) in the core network. The proxy binding update process is initiated, and through the cooperation of T-MA, AME, and SME, the IP address assigned to the terminal is forwarded to the mobile terminal MN, and the MN accesses the T-MA. Then, the MN sends a service request packet to the T-MA, which contains the IP address assigned by the S-MA (SourceMA, source mobility anchor). T-MA sends the service data packet to S-MA according to the source IP address in the data packet. After receiving the data packet, S-MA sends the data packet to S-MEP (SourceMobile Edge Platform) located in the S-MA service range. , that is, the source mobile edge platform), the S-MEP sends it to the corresponding S-APP, and at the same time sends an APP relocation request to the MEO (Mobile Edge Orchestrator, that is, the mobile edge orchestrator). Next, MEO, MEPM (Mobile Edge Platform Manager), and MEP collaborate to perform APP migration management. After the management work is completed, S-MEP, S-MEPM, MEO, T-MEP, and T-MEPM collaborate to migrate the APP instance and update the routing rules in T-MEP. Afterwards, T-MEP activates the migrated APP (T-APP) for the terminal node MN. If necessary, S-MEP terminates the S-APP instance and releases resources. After the MEC service relocation is completed, the MN communicates with T-APP through T-MA, T-MEP and T-APP. In this solution, APP is always in the same service range as MN, ensuring the low latency and high reliability of MEC services. However, when the MN moves across gateways, the MN leaves the S-MA service range and then performs MA switching and APP migration in sequence, which cannot support business continuity, causing edge business interruption.

In ETSI solution 2, the MN leaves the service range of S-MA and moves to the service range of T-MA. First, MN performs signal detection, and the MA with the strongest detected signal is the T-MA. Then the MN sends a routing request to the T-MA. After the MA receives the routing request, the T-MA initiates a proxy binding update process with the AME and SME in the core network. Through the cooperation of the T-MA, AME and SME, the assigned The IP address of the terminal is sent to the terminal node MN through the router, and the MN accesses the T-MA. After the MA migration is completed, the MN sends the MEC service packet to the T-MA, which contains the source IP address. The T-MA forwards the service packet to the S-MA based on the source IP address. The S-MA forwards the message to the S-MEP located in the service range, and the S-MEP sends the service message to the corresponding S-APP. Afterwards, S-MEP, S-APP, S-MEPM, and MEO collaborate to perform business management work and update the PDU session information between the MN and S-APP, including session ID, IP address, traffic rules, etc. After the management work is completed, S-MEP activates a new session for the MN through S-MA and T-MA. The MEC service relocation is completed, and the MN communicates with S-APP through T-MA, S-MA, S-MEP and S-APP. Compared with Solution 1, the APP has not been migrated, which reduces the time for MEC service relocation and supports the business continuity when MN switches across gateways to a certain extent. However, this solution comes at the expense of a longer roundabout forwarding path, leading to many problems such as delay, reliability, and route management.

In SSC mode 1 of 3GPP, when the MN moves across gateways, the business services provided by the network to the MN will not be dropped. During the session, the network will always maintain the initial mobility anchor point MA, and will not be affected by the MN's mobility or access to the network. The IP address assigned to the MN will not change as the technology changes. This mode can be applied to any session and any access type. However, this model comes at the expense of a longer roundabout forwarding path, resulting in many problems such as delay, reliability, and routing management.

In SSC mode 2 of 3GPP, when the MN moves across gateways, the network releases the connection service with the MN and the corresponding application session, and the originally assigned IP address will also be released. When the mobility anchor point changes in SSC mode 2, the original connection is first released, then the connection between the MN and the new mobility anchor point is established, and the IP address is re-allocated to the MN. This model cannot meet business continuity and will cause business continuity interruption.

In SSC mode 3 of 3GPP, when the MN moves across gateways and the mobility anchor changes, a connection with the new session anchor is established before the change, and the new connection also accesses the same data network DN. After the new mobility anchor allocates a new IP address to the MN, it instructs the MN to maintain the old IP address for a period of time through NAS (Network Attached Server) signaling, and then releases the old IP address. Session resources are also released accordingly. In this mode, the user's communication address will change as the user's anchor point changes, but since a new connection has been established before, the business will not be interrupted. However, after this mode is completed, the MN still conducts communication sessions with the same DN, and the data forwarding path is circuitous and lengthy, causing many problems such as delay, reliability, and routing management.

Therefore, when the MN moves across gateways, the above solution cannot guarantee service continuity and low-latency and high-reliability service quality. The main problems are as follows:

(1) Existing MEC business relocation solutions provide multiple types of session and business continuity modes, such as the three SSC modes of 3GPP, to meet various continuity requirements of different applications and services. The process of such a solution is cumbersome and the network deployment is complicated. How to implement a unified MEC service relocation process to cope with various continuity requirements of different scenarios and services is not clear in the existing technology.

(2) When the MN moves across gateways, service relocation requires network layer MA switching and application layer APP migration in sequence. Service communication can be resumed only after the service relocation is completed. The processing time is too long and exceeds the interruption time range that the business can tolerate, causing interruption of business continuity. The non-migration of APPs ensures business continuity to a certain extent, but the forwarding path is circuitous and lengthy, causing problems such as delay, reliability, and routing management.

(3) The APP instance identifies the identity of the mobile terminal based on the unified communication address (such as IP address) of the identity location and some APPs do not support the change of the communication address of the mobile terminal MN during the session, so the mobility anchor point is fixed. When the MN moves across gateways, the APP migrates. However, the mobility anchor point is fixed, the data packet forwarding path is circuitous, and the process is complex, causing problems such as delay, reliability, and routing management.

(4) When the MN moves across gateways, the connecting node MA switches, and the APP needs to be moved to a location close to the user. The existing technology is not clear on how to achieve notification collaboration between the network and the application and the consistency between the network layer connection node and the target application location.

(5) Mobile terminal devices are diverse, such as self-driving cars, drones, satellites, ocean-going freighters, etc. The existing paradigm of combining unified identity and location communication addresses with mobility anchor points cannot effectively support multi-mode access, which increases the complexity of network deployment and device access. Therefore, a new solution is urgently needed to solve the above technical problems.

Contents of the invention

The present invention is aimed at the problems existing in the existing technology and provides a relocation method to ensure MEC business continuity. This technical solution adopts a unified MEC business relocation method to adapt to various continuity requirements of different scenarios and services. , achieving lower service migration overhead in exchange for smaller service delay, while simplifying the MEC service relocation process and network deployment.

In order to achieve the above objects, the technical solution of the present invention is as follows, a relocation method to ensure MEC business continuity, the method includes the following steps:

Step S1: The mobile terminal MN goes online, completes the registration and session registration process, and binds the session to the terminal identity;

Step S2: The mobile terminal MN is about to undergo cross-gateway handover, triggering the network layer and application layer notification coordination mechanism, and the application layer senses the network layer status change;

In step S3, the mobile terminal MN undergoes cross-gateway handover, and collaboratively performs network layer handover management and application layer edge application migration process.

The details of step S1 are as follows. The mobile terminal MN registration and session registration process are as follows:

Step S101: The mobile terminal MN goes online, has a unique user identity identifier (UID) in the entire network, sends a routing request to its access service node (ASN) according to the signal strength, and requests access to the ASN;

Step S102: After receiving the routing request from the MN, the ASN sends an authentication and authorization request to the access and mobility management network element (AME) in the core network. The AME performs authentication services for the terminal requesting access and determines access. Whether the terminal is legal, AME authorizes the legal terminal; ASN allocates an appropriate routing identifier (RID) to the authorized MN, and records the MN's UID/RID mapping information into the local cache;

Step S103, the ASN sends a UID/RID mapping information update request to the Identity Location Register (ILR) to inform the ILR of the latest mapping information of the mobile terminal;

Step S104: After receiving the update of the mapping information, the ILR records the mapping information UID/RID of the mobile terminal into the local cache so that subsequent communication peers can initiate communication to the MN to query the routing information of the MN;

Step S105: After the access work is completed, the ASN sends a routing request success response to the MN. The MN client application (APP) establishes a session to the edge APP in the edge data center. This session includes multiple types, such as a communication data unit (PDU) session in the 5G network and a packet data network (PDN) session in the 4G network. First, the MN sends a session establishment request to the AME in the core network through the ASN;

Step S106: The AME selects the session management network element SME for the session that the MN needs to establish based on the MN's RID information;

Step S107: The AME sends an MN session creation request to the selected session management network element (SME). After receiving the request, the SME feeds back response information to the AME;

Step S108, AME and SME collaborate to authenticate the session to be established, determine whether the session to be established is legal, and authorize the legal session;

Step S109: The SME sends an MN session creation request to the ASN accessed by the MN. After receiving the session request, the ASN feeds back response information to the SME;

Step S110: After receiving the response information, the SME forwards the session information to be created to the AME. After receiving the response information, the AME feeds back the response information to the SME;

Step S111, the ASN establishes an ASN-specific session with the mobile terminal through the access network (RAN). The session identifier is the user identity identifier UID of the MN. The establishment process includes sending session request response information to the mobile terminal MN.

At this point, the session establishment between the MN and the mobile edge application (ME APP) is completed.

Among them, in step S2, the network layer and application layer notify the coordination mechanism, as follows:

Operators and third-party applications can implement a collaborative mechanism for mutual notification through messages between Policy Control Elements (PCE), Application Functions (AF), and Network Capability Exposure Network Elements (NEE), thereby enabling third-party applications to be based on business characteristics. Send requirements to the network and promptly notify applications when the MN location and network status change.

Step S201, the application layer subscribes to network event notifications, which can be implemented through AF in three ways:

Method 1: The application layer transmits the application's request for network resources through the AF request message, thereby realizing the network layer's delay and bandwidth requirements required for the service.

Method 2: The application layer subscribes to event notification messages to the network layer through AF/NEE. For example, when a node switches across gateways, AF is notified immediately when the network detects that the subscribed event is triggered.

Method 3: The application layer initiates the modification process of terminal data routing through AF, and adjusts the uplink and downlink and multi-homing policies to manage the access connection point ASN.

Step S202: After receiving the subscription message, the AF sends an AF request affecting the routing policy control information to the PCE;

Step S203: The network layer PCE receives the AF subscription message. For example, the Radio Network Information Service (RNIS) senses that the access connection point ASN of the session has changed, the data network (DN) access identifier (DNAI) has changed, and other events occur. The network layer will send a notification message to the application layer, including the changed target data network identifier DNAI. The purpose is for the application layer to obtain the target ME host location for application migration and ensure the consistency of the access node ASN and the target APP location.

Among them, the network layer switching management and application layer edge application migration process described in step S3 are as follows:

The mobile terminal MN is about to leave the service range of the source access service node S-ASN and performs signal detection. It detects that the signal of the new ASN is greater than the signal of the S-ASN, and the difference in signal strength exceeds a certain threshold, and sets this ASN as the destination. Access the service node T-ASN, and the mobile application platform of T-ASN serves as the target mobile application platform T-MEP, triggering the migration of the application layer edge application MEAPP.

When the mobile terminal MN moves across gateways, the application layer edge application ME APP migration process is divided into 6 modules.

Module 1, MN and S-ASN separation,

Step S301: The MN is about to leave the service range of the S-ASN. The client APP in the MN sends an application move request to the S-MEP in the source edge network, and the S-MEP feeds back a response to the MN;

Step S302, S-MEP enables application mobility to prepare for S-APP migration, which includes locating the S-APP that is in session with the MN APP, organizing session context data, etc.;

Module 2, MN accesses T-ASN.

Step S303: The MN network layer performs handover management and application layer session update process;

Module 3, update traffic rules for MN,

Step S304: The mobile terminal MN accesses the T-ASN. The S-RNIS receives the handover message of the MN and sends the MN area change notification to the S-MEP. After receiving the message, the S-MEP feeds back a confirmation response to the S-RNIS;

Step S305, S-MEP queries S-MEPM for the target host T-MEC host that allows the MN to receive information. S-MEPM forwards the query to MEO. MEO passes the operator policy, mobility policy, application requirements, application capabilities, and mobility A series of strategic methods such as edge system status, MEO finally determines the T-MEC host and forwards the information to S-MEP through S-MEPM so that S-MEP can contact T-MEP. At the same time, S-MEP obtains the need to migrate in the platform S-APP and session context information to T-MEP;

Step S306, S-MEP queries the affected traffic routing rules, which include MNUID as part of the service filter and application information as part of the application program;

Step S307, S-MEP sends a traffic routing rule update request to T-MEP, which contains application information and affected traffic routing rules;

Step S308, the T-MEP receives the traffic routing rule update request from the S-MEP and updates the local traffic routing rules;

Step S309: After the update is completed, feed back the traffic routing rule update response to the S-MEP;

Module 4, Application Migration Management,

Step S310: The MN network layer switching is completed, the T-MEP routing rule update is completed, and the MN APP sends an application migration request to the S-MEP;

Step S311, S-MEP captures the status information of the serving S-APP and obtains the completed session context information in step S305;

Step S312, S-MEP sends an application migration request to MEO through S-MEPM, including the S-APP ID and the ID of the target node;

Step S313, MEO sends an instantiation application request to T-MEPM, and T-MEPM instantiates T-APP on the target T-MEC host;

Step S314: The instantiation is completed, and MEO feeds back the application migration response to S-MEP through S-MEPM;

Step S315, the S-MEP sends an application status information transmission request to the T-MEP. The application status information transmission request includes the captured status information and session context information of the S-APP;

Step S316, T-MEP synchronizes the received service status information and context information with the T-APP running on the T-MEC host, and activates it;

Step S317, T-MEP sends an application status information transmission response to S-MEP to indicate whether the service status transmitted by the application instance is successful;

Step S318: After the service status information and context information are synchronized, the T-APP instance sends an application instance running notification to the T-MEP;

Step S319, T-MEP forwards the T-APP application instance running notification to S-MEP;

Step S320, S-MEP sends an application migration response to the MN APP to trigger the next module;

Module 5, activate new traffic routing rules,

Step S321, T-APP activates traffic routing rules to T-MEP;

Step S322, T-MEP activates traffic routing rules;

Step S323: T-MEP forwards the traffic routing rule activation notification to S-MEP;

Step S324, S-APP sends a request to deactivate or update local traffic rules to S-MEP;

Step S325, S-MEP receives the RNIS area change notification from S-RNIS;

Step S326, when S-MEP receives the traffic routing rule activation notification from T-MEP and the traffic routing deactivation or update request from S-APP, S-MEP deactivates or updates the corresponding traffic routing rules;

Module 6, terminate the source application.

Step S327: If it is necessary to save resources, etc., S-MEP, S-MEPM, MEO, S-APP, and S-RNIS collaboratively terminate the source application.

Among them, step S303 network layer switching management process and application session update process are as follows:

Step S303-1: MN triggers network switching and sends a session modification request to AME, and the request message contains the session ID;

Step S303-2: AME receives the session modification request and sends the session modification request to the session management network element SME corresponding to the session according to the session ID. After receiving the request, SME feeds back a session modification response to AME to prepare for the session modification;

Step S303-3: After receiving the response, the AME sends a session modification command to the MN, and the MN starts the timer and performs network switching;

Step S303-4: The MN initiates a routing request to the T-ASN. The request contains authorization information and requests access to the T-ASN;

Step S303-5: T-ASN receives the routing request, assigns a new routing identifier RID2 to the MN, and records the UID/RID2 mapping information into the local cache;

Step S303-6: T-ASN establishes a temporary transmission tunnel to S-ASN. At the same time, T-ASN sends a mapping information update request to ILR. After receiving the update request, ILR updates the mapping information in the local cache and records the latest MN. mapping information;

Step S303-7: T-ASN sends a successful response to the routing request to the MN and performs subsequent session modification work;

Step S303-8: After the access work is completed, the MN sends a new session establishment request to the AME through the T-ASN, and the data packet contains the IDs of the old and new sessions;

Step S303-9: After receiving the new session establishment request, the AME sends a session modification request to the SME that manages the old session, including the old and new session IDs;

Step S303-10: The SME that manages the old session determines the ASN that needs to be executed and relocates the ASN;

Step S303-11: Feed back the AME session modification response. This data packet contains the information of the new session;

Step S303-12: SME sends a session modification request to T-ASN, and T-ASN feeds back a session modification response to SME;

Step S303-13, T-ASN establishes a T-ASN-specific session with the MN, and sends a new session establishment success response to the MN after the establishment is completed;

Step S303-14: After the new session is established, the wait timer stops and the MN releases the session resources with the S-ASN.

A structure of a relocation method that ensures MEC business continuity, the structure includes a network layer structure and an application layer structure,

The network layer structure is divided into central DC (Data Center, i.e. data center), local DC, and edge DC. Application resources are offloaded to locations closer to users. Mobile terminals include multiple types. For example, mobile phones, self-driving cars, high-speed trains, drones, etc. The relocation network layer structure to ensure MEC business continuity according to the present invention is shown in Figure 1, and the application layer structure is shown in Figure 2, in which the main network elements and functional entities include:

Among them, the network layer structure includes

Access service node ASN, ASN is the interface between the access network and the core network, and is also the access connection point of the mobile terminal MN. It is responsible for ensuring the communication connection between the MN and the entire network, assigning the location address RID to the MN, registration, authentication, and authorization. , Traffic accounting, caching the UID/RID mapping relationship of the mobile terminal at the control level. In the data level, ASN is responsible for encapsulating and decapsulating data packets in the network.

Identity Location Register (Identifier Locator Register), ILR is responsible for maintaining and managing the latest UID/RID mapping relationship of the mobile terminal, the initial registration process, and the RID query of the WAN communication peer.

Session management network element SME, SME is responsible for session management, tunnel maintenance, route identification allocation and management, ASN selection, policy implementation, QoS control and traffic accounting. This network element includes many types, such as SMF (SessionManagement Function) in 5G network , that is, the session management function), the session management function of the MME in the 4G network, etc.

Access and mobility management network element AME. AME is the access point of terminals and wireless core networks. It is responsible for performing registration, connection, reachability, and mobility management, and provides session management information transmission channels for mobile terminals and SMEs. Terminal access provides authentication and authentication functions. This network element includes various types, such as AMF (Access and Mobility Management Function) in the 5G network, and the NAS access control function in the MME in the 4G network.

Policy Control Network Element PCE (Policy Control Network Element), PCE supports a unified policy framework for controlling network behavior, provides policy rules for control plane execution, and accesses subscription information related to policy formulation in the database. This network element It includes many types, such as PCF (Policy Control Function) in 5G network and PCRF (Policy and Charging Rule Function) in 4G network.

Network capability exposure network element NEE (Network Exposure Network Element), NEE provides external exposure of network functional capabilities. External exposure can be divided into monitoring capabilities, supply capabilities, traffic routing applications and impacts, and policies/planning. spending ability. This network element includes many types, such as NEF (Network Exposure Function, that is, network capability exposure) in the 5G network, and SCEF (Service Capability Exposure Function, that is, service capability exposure) in the 4G network.

Among them, the application layer structure includes:

Mobile edge application ME APP. ME APP is an application instance running on the virtualized infrastructure. It can interact with the mobile edge platform to obtain the service-oriented open capabilities of the mobile edge platform.

Mobile edge platform MEP, MEP is responsible for providing mobile edge services for APP, including: service registration, service discovery, status monitoring, local offloading, DNS service, load balancer, firewall, as well as wireless network information service, location information service, and bandwidth management service and a series of wireless network capability services. In the collaboration mechanism of the distributed MEC system, it can be interconnected with different MEPs.

Radio Network Information Service RNIS (Radio Network Information Service), RNIS provides wireless network related information services to ME APP. ME APP can obtain the desired wireless network information by consulting or subscribing, such as mobile The terminal switches.

Mobile edge platform manager MEPM. MEPM is responsible for the basic operation and maintenance of MEP, mobile edge service configuration, life cycle management of ME APP, and application rules and demand management of ME APP.

Mobile edge orchestrator MEO, MEO is the orchestration center of MEC services. Usually only one is deployed in the country, located in the central DC. MEO macroscopically controls all resources and capacities of the MEC platform, mainly including computing, storage, and network resources in the virtual infrastructure manager. , application mirror resources, check the integrity and authenticity of the software package, and then also need to measure the user resource requirements and the available resources of each ME host, and select the most appropriate ME host for deployment.

Application Function AF (Application Function) refers to various services in the application layer of the core network. It can be an operator's internal application or a third-party AF such as a video server. It can also be placed at the edge of the network, such as RNIS.

Data forwarding plane software DP (Data Plane), DP is responsible for executing the traffic rules issued by the MEP and processing the traffic between MEAPP, ME service, DNS server, proxy server, 3GPP network, other access networks, LAN and external networks.

Compared with the existing technology, the present invention has the following advantages: (1) It adopts a unified MEC service relocation method to adapt to different scenarios and various continuity requirements of services, and achieves lower service migration overhead in exchange for smaller services. delay, while simplifying the MEC service relocation process and network deployment.

(2) The session information between applications is bound to the identity of the mobile terminal. The identity is unique in the entire network, which reduces the impact of changes in the unified communications address of the identity location on the terminal service delay reliability and solves the problem that some MEC applications do not support sessions. The communication address changes during the process, while ensuring the continuity of MEC service relocation when the mobile terminal moves across gateways.

(3) Using the above network identification method, data packet forwarding does not involve mobility anchor points. After cross-gateway movement, the access connection point and the APP are at the same location close to the user. On the basis of ensuring business continuity, routing detours are eliminated. , reducing unnecessary node and link delays and improving forwarding reliability.

(4) The network notifies the application platform when it senses the location of the application, carries out the preparation process in advance, and quickly allocates new application instances on the destination MEP through collaboration between the application platform and the network. It accelerates the application's perception of terminal behavior and network status, achieves location consistency between access connection points and target applications, and enables rapid application migration and business continuity guarantee of MEC services during mobile handover.

(5) Identity identifiers and routing identifiers are used instead of unified network addresses, and network forwarding does not involve mobility anchors. This paradigm simplifies the network architecture and network resources, and supports easy access for multiple types of terminals.

(6) In this solution, the network administrator assigns a unique user identity identifier (UID) to each mobile terminal device, the access service node assigns a routing identifier (RID) to the mobile terminal, and the UID/RID mapping information replaces the original unified identity location. Communication address, the mobile edge platform binds the application session to the mobile terminal identity UID.

(7) Collaborative migration of mobile terminals, network connections and edge applications: When the mobile terminal is about to move across gateways, the mobile terminal, network connection and edge applications collaborate to migrate across gateways.

(8) The network layer and application layer structure of the mobility anchor edge network are not involved: the edge network network layer consists of mobile terminals, access service nodes and data networks, and the application layer structure consists of client applications, access service node applications, It consists of data plane software, mobile edge platform, radio network information services, and mobile edge applications. No mobility anchors are involved in the network, routing detours are eliminated, unnecessary node delays and link delays are reduced, and the reliability of data packet forwarding is improved. Mobile terminal online registration and session registration process, the session is bound to the terminal identity.

(9) In this solution, when a mobile terminal switches across gateways, the network layer and application layer notify the coordination mechanism: when the network senses that a terminal switching event occurs, it notifies the application platform, carries out the preparation process in advance, and speeds up the application's perception of terminal behavior and network status. Achieve location consistency between access connection points and target applications.

(10) Application layer edge application migration signaling process when mobile terminals switch across gateways: Provide a unified and efficient application layer edge application migration method to simplify the application migration process and reduce business migration delays.

(11) When a cross-gateway handover occurs in a mobile terminal, the signaling process of network layer handover management and application layer session update: provides an efficient network layer cross-gateway handover method to reduce handover time.

(12) A process for forwarding data packets between mobile terminals and WAN application hosts: providing a data packet forwarding process that does not involve mobility anchor points for mobile terminals and WAN application hosts.

(13) A deployment scheme and operation mechanism of a network control plane ILR based on DHT-MAP: The control plane consists of several ILRs, divided into different autonomous domains. The ILRs in each domain store and query terminal mapping information based on DHT-MAP.

Description of drawings

Figure 1 is a schematic diagram of the network layer structure and the data packet forwarding path during the handover process of the relocation method to ensure MEC business continuity according to the present invention;

Figure 2 is a schematic diagram of the application layer structure and data transmission path of the switching process of the relocation method to ensure MEC business continuity according to the present invention;

Figure 3 is a schematic diagram of the application layer edge application migration process when cross-gateway switching occurs in the mobile terminal of the present invention;

Figure 4 is a schematic diagram of the flow of network layer handover management and application session update when cross-gateway handover occurs in the mobile terminal of the present invention;

Figure 5 is a schematic diagram of the network layer and application layer notification coordination mechanism when cross-gateway switching occurs in the mobile terminal of the present invention;

Figure 6 is a schematic diagram of the data packet forwarding structure between the mobile terminal and the wide area network application host according to the present invention.

Detailed ways

In order to deepen the understanding of the present invention, this embodiment will be described in detail below with reference to the accompanying drawings.

Embodiment 1: The network layer and application layer structures of the relocation method to ensure MEC business continuity are shown in Figures 1-2.

The network layer structure of the relocation method for ensuring MEC business continuity according to the present invention is shown in Figure 1. The network layer is divided into a central DC (Data Center), a local DC, and an edge DC. Application resources are offloaded to A location closer to the user. Mobile terminals include many types, such as mobile phones, self-driving cars, high-speed trains, drones, etc. The relocation network layer structure to ensure MEC business continuity according to the present invention is shown in Figure 1, and the application layer structure is shown in Figure 2, in which the main network elements and functional entities include:

Network layer:

Access service node ASN. ASN is the interface between the access network and the core network, and is also the access connection point of the mobile terminal MN. Responsible for ensuring the communication connection between MN and the entire network, assigning location address RID to MN, registration, authentication, authorization, traffic accounting, etc. The UID/RID mapping relationship of the mobile terminal is cached at the control level. In the data plane, ASN is responsible for encapsulating and decapsulating data packets in the network.

Identity location register ILR (Identifier Locator Register). The ILR is responsible for maintaining and managing the latest UID/RID mapping relationship of the mobile terminal, the initial registration process, and the RID query of the WAN communication peer.

Session management network element SME. SME is responsible for session management, tunnel maintenance, route identification allocation and management, ASN selection, policy implementation, QoS control and traffic accounting, etc. This network element includes many types, such as SMF (Session Management Function) in the 5G network, the session management function of the MME in the 4G network, etc.

Access and mobility management network element AME. AME is the access point for terminals and wireless core networks and is responsible for performing registration, connection, reachability, and mobility management. Provides session management information transmission channels for mobile terminals and SMEs, and provides authentication and authentication functions for mobile terminal access. This network element includes many types, such as AMF (Access and Mobility Management Function) in the 5G network, and the NAS access control function in the MME in the 4G network.

Policy Control Network Element PCE (Policy Control Network Element). PCE supports a unified policy framework for controlling network behavior, provides policy rules for control plane execution, and accesses subscription information related to policy formulation in the database. This network element includes many types, such as PCF (Policy Control Function) in the 5G network and PCRF (Policy and Charging Rule Function) in the 4G network.

Network capability exposure network element NEE (Network Exposure Network Element, that is, network capability exposure network element). NEE provides external exposure of network functional capabilities. External exposure can be divided into monitoring capabilities, provisioning capabilities, application and impact of traffic routing, and policy/billing capabilities. This network element includes many types, such as NEF (Network Exposure Function, that is, network capability exposure) in the 5G network, and SCEF (Service Capability Exposure Function, that is, service capability exposure) in the 4G network.

Application layer:

Mobile edge application ME APP. ME APP is an application instance running on virtualized infrastructure and can interact with the mobile edge platform to obtain the service-oriented open capabilities of the mobile edge platform.

Mobile edge platform MEP. MEP is responsible for providing mobile edge services for APPs, including: service registration, service discovery, status monitoring, local offloading, DNS services, load balancers, firewalls, and a series of wireless networks such as wireless network information services, location information services, and bandwidth management services. Ability to serve. In the collaboration mechanism of the distributed MEC system, it can be interconnected with different MEPs.

Radio Network Information Service RNIS (Radio Network Information Service). RNIS provides wireless network related information services to ME APP. ME APP can obtain the desired wireless network information by querying or subscribing, such as when a mobile terminal switches.

Mobile Edge Platform Manager MEPM. MEPM is responsible for the basic operation and maintenance of MEP, mobile edge service configuration, life cycle management of ME APP, and application rules and demand management of ME APP.

Mobile edge orchestrator MEO. MEO is the orchestration center of MEC services. Usually there is only one deployed in the country, located in the central DC. MEO macroscopically masters all resources and capacities of the MEC platform, mainly including computing, storage, network resources, application image resources in the virtual infrastructure manager, checking the integrity and authenticity of software packages, and then also needs to measure user resource requirements and Based on the available resources of each ME host, select the most appropriate ME host for deployment.

Application function AF (Application Function). AF refers to various services in the application layer of the core network, which can be internal applications of operators or third-party AF such as video servers. It can also be placed at the edge of the network, such as RNIS.

Data forwarding plane software DP (Data Plane). The DP is responsible for executing the traffic rules issued by the MEP and processing the traffic between MEAPP, ME service, DNS server, proxy server, 3GPP network, other access networks, LAN and external networks.

Further, the present invention introduces two identifiers for the user terminal, a user identity identifier (User Identifier, referred to as UID) and a routing identifier (Routing Identifier, referred to as RID). UID is a unique identifier assigned to mobile terminals by the network administrator. It represents the identity of the node and remains static throughout the network and cannot be changed. Mainly responsible for identifying the identity of the terminal when communicating between the terminal and the application, and between the terminal and the communication peer. RID is used to represent the current location of the terminal host and indicates the routing direction for data traffic during terminal communication. UID/RID mapping information is used to replace the communication address with unified identity and location, and the mobile edge application platform binds the application session to the UID of the mobile terminal MN. The edge DN communicates through UID, and the WAN communicates through RID. The access network is separated from the core network, the connection point is ASN, and the mobility anchor is not involved in network forwarding.

In the MEC relocation method proposed by the present invention, a unified MEC relocation method is provided to cope with different scenarios and various continuity requirements of services. The collaborative migration of mobile terminals, network connections and edge applications ensures the continuity of MEC services.

Embodiment 2: A relocation method to ensure MEC business continuity. The method includes the following steps: The method includes the following steps:

Step S1: The mobile terminal MN goes online, completes the registration and session registration process, and binds the session to the terminal identity;

Step S2: The mobile terminal MN is about to undergo cross-gateway handover, triggering the network layer and application layer notification coordination mechanism, and the application layer senses the network layer status change;

In step S3, the mobile terminal MN undergoes cross-gateway handover, and collaboratively performs network layer handover management and application layer edge application migration process.

details as follows:

In step S1, the mobile terminal MN registration and session registration process:

Taking the mobile terminal MN as an example, its registration and session registration steps are:

Step S101: The mobile terminal MN goes online, sends a routing request to the ASN to which it belongs based on the signal strength, and requests access to the ASN;

Step S102: After receiving the routing request from the MN, the ASN sends an authentication and authorization request to the AME in the core network. The AME performs authentication services for the terminal requesting access, determines whether the access terminal is legal, and the AME authorizes the legal terminal;

Step S103, the ASN allocates the appropriate routing identifier RID to the authorized MN, and records the MN's UID/RID1 mapping information into the local cache;

Step S104, the ASN sends a UID/RID1 mapping information update request to the ILR to inform the ILR of the latest mapping information of the mobile terminal;

Step S105: After receiving the update of the mapping information, the ILR records the mapping information UID/RID1 of the mobile terminal into the local cache so that subsequent communication peers can initiate communication to the MN to query the routing information of the MN;

Step S106: After the access work is completed, the ASN sends a routing request success response to the MN. The MN client APP establishes a session to the edge APP in the edge data center. This session includes multiple types, such as PDU sessions in the 5G network and PDN sessions (Packet Data Network) in the 4G network. First, the MN sends a session establishment request to the AME in the core network through the ASN;

Step S107: The AME selects the session management network element SME for the session that the MN needs to establish based on the MN's RID information;

Step S108: AME sends an MN session creation request to the selected SME, and after receiving the request, the SME feeds back response information to the AME;

Step S109: AME and SME collaborate to authenticate the session to be established, determine whether the session to be established is legal, and authorize the legal session;

Step S110: The SME sends an MN session creation request to the ASN accessed by the MN. After receiving the session request, the ASN feeds back response information to the SME;

Step S111: After receiving the response information, the SME forwards the session information to be created to the AME. After receiving the response information, the AME feeds back the response information to the SME;

Step S112: The ASN establishes an ASN-specific session with the mobile terminal through the access network RAN. The establishment process includes sending session request response information to the mobile terminal MN.

After the session between the MN and the ME APP is established, the network layer data packet forwarding path is path a in Figure 1; the application layer data communication path is path a in Figure 2.

In step S2, the mobile terminal MN switches network layer and application layer notification coordination mechanism across gateways.

The relocation mechanism to ensure MEC business continuity requires the application layer to support cross-gateway application migration and fast context migration, and also requires a notification coordination mechanism between the network layer and the application layer.

Operators and third-party applications can implement a collaborative mechanism for mutual notification through messages between policy control network elements PCE, application functions AF, and network capability opening network elements NEE, as shown in Figure 5, thereby enabling third applications to be based on business characteristics. Send requirements to the network and promptly notify applications when the MN location and network status change.

Step S201: The application layer subscribes to network event notifications. This can be achieved with AF in 3 ways.

Method 1: The application layer transmits the application's request for network resources through the AF request message, thereby realizing the network layer's delay and bandwidth requirements required for the service.

Method 2: The application layer subscribes to event notification messages to the network layer through AF/NEE, such as a node switching across gateways, so that when the network detects that the subscribed event is triggered, AF is notified immediately.

Method 3: The application layer initiates the modification process of terminal data routing through AF, and adjusts the uplink and downlink and multi-homing policies to manage the ASN of the access connection point.

Step S202: After receiving the subscription message, the AF sends an AF request affecting the routing policy control information to the PCE;

Step S203: The network layer PCE receives the AF subscription message. For example, the RNIS senses that the access connection point ASN of the session has changed or the DN access identifier DNAI (Data Network Access Identifier) has changed. The network layer will send a notification message. To the application layer, include the changed target data network identifier DNAI. The purpose is for the application layer to obtain the location of the target ME host for application migration and ensure the consistency of the access node ASN and the location of the target APP;

AF initiates application cross-gateway migration and synchronizes application status information and session context information to the ME APP in the target ME host. During application migration, the network side forwards data packets with the source DN through a temporary tunnel to keep services uninterrupted. After the application migration is completed, the AF notifies the SME that the migration is completed, and the SME initiates a data transmission path update to forward the client services in the mobile terminal from T-ASN to the target mobile edge application T-APP, achieving seamless cross-gateway switching mobility. manage.

In step S3, the mobile terminal MN undergoes a cross-gateway switching application layer edge application migration process.

The following takes the cross-gateway handover of the mobile terminal MN as an example to illustrate the application layer edge application migration management process during the movement of the mobile terminal, as shown in Figure 3.

The mobile terminal MN is about to leave the service range of the source access service node S-ASN and performs signal detection. It detects that the signal of the new ASN is greater than the signal of the S-ASN, and the difference in signal strength exceeds a certain threshold, and sets this ASN as the destination. Access the service node T-ASN, and the mobile application platform of T-ASN serves as the target mobile application platform T-MEP, triggering the migration of the application layer edge application MEAPP.

The mobile terminal MN moves across gateways, and the application layer edge application ME APP migration process is divided into six modules.

Module 1, MN and S-ASN are separated.

Step S301: The MN is about to leave the service range of the S-ASN. The client APP in the MN sends an application move request to the S-MEP in the source edge network, and the S-MEP feeds back a response to the MN;

Step S302, S-MEP enables application mobility to prepare for S-APP migration, which includes locating the S-APP that is in session with the MN APP, organizing session context data, etc.;

Module 2, MN accesses T-ASN.

Step S303, MN network layer switching management and application session update process;

Module 3, updates traffic rules for the MN.

Step S304: The mobile terminal MN accesses the T-ASN. The S-RNIS receives the handover message of the MN and sends the MN area change notification to the S-MEP. After receiving the message, the S-MEP feeds back a confirmation response to the S-RNIS;

Step S305: S-MEP queries S-MEPM for the target host T-MEC that allows the MN to receive the information, and S-MEPM forwards the query to the MEO. MEO uses a series of strategies such as operator strategy, mobile strategy, application requirements, application capabilities, and mobile edge system status. Finally, MEO determines the T-MEC host and forwards the information to S-MEP through S-MEPM so that S-MEP Ability to contact T-MEP. At the same time, S-MEP obtains the S-APP and session context information in the platform that needs to be migrated to T-MEP;

Step S306, S-MEP queries the affected traffic routing rules, which include MNUID as part of the service filter and application information as part of the application program;

Step S307, S-MEP sends a traffic routing rule update request to T-MEP, which contains application information and affected traffic routing rules;

Step S308, the T-MEP receives the traffic routing rule update request from the S-MEP and updates the local traffic routing rules;

Step S309: After the update is completed, feed back the traffic routing rule update response to the S-MEP;

Module 4, application migration management.

Step S310: The MN network layer switching is completed, the T-MEP routing rule update is completed, and the MN APP sends an application migration request to the S-MEP;

Step S311, S-MEP captures the status information of the serving S-APP and obtains the completed session context information in step 305;

Step S312, S-MEP sends an application migration request to MEO through S-MEPM, including the S-APP ID and the ID of the target node;

Step S313, MEO sends an instantiation application request to T-MEPM, and T-MEPM instantiates T-APP on the target T-MEC host;

Step S314: The instantiation is completed, and MEO feeds back the application migration response to S-MEP through S-MEPM;

Step S315, the S-MEP sends an application status information transmission request to the T-MEP. The application status information transmission request includes the captured status information and session context information of the S-APP;

Step S316, T-MEP synchronizes the received service status information and context information with the T-APP running on the T-MEC host, and activates it;

Step S317, T-MEP sends an application status information transmission response to S-MEP to indicate whether the service status transmitted by the application instance is successful;

Step S318: After the service status information and context information are synchronized, the T-APP instance sends an application instance running notification to the T-MEP;

Step S319, T-MEP forwards the T-APP application instance running notification to S-MEP;

Step S320, S-MEP sends an application migration response to the MN APP to trigger the next module;

Module 5, activate new traffic routing rules.

Step S321, T-APP activates traffic routing rules to T-MEP;

Step S322, T-MEP activates traffic routing rules;

Step S323: T-MEP forwards the traffic routing rule activation notification to S-MEP;

Step S324, S-APP sends a request to deactivate or update local traffic rules to S-MEP;

Step S325, S-MEP receives the RNIS area change notification from S-RNIS;

Step S326, when S-MEP receives the traffic routing rule activation notification from T-MEP and the traffic routing deactivation or update request from S-APP, S-MEP deactivates or updates the corresponding traffic routing rules;

Module 6, terminate the source application.

Step S327: If it is necessary to save resources, etc., S-MEP, S-MEPM, MEO, S-APP, and S-RNIS collaboratively terminate the source application.

The data transmission path between MN APP and S-APP before MN moves across gateways is shown as path a in Figure 2. MN APP transmits data through S-DP and S-APP in S-MEP; during the application migration process, MN APP transmits data with S-APP in S-MEP through the temporary tunnel between T-DP and S-DP. The path is shown as path b in Figure 2; after the APP migration is completed, MNAPP communicates with T-DP through T-DP. The T-APP in the MEP directly transmits data, as shown in path c in Figure 2.

In step 303, the mobile terminal MN undergoes a cross-gateway handover network layer handover management and application layer session update process.

The following takes the cross-gateway handover of the mobile terminal MN as an example to illustrate the network layer handover management process and application layer session update during the movement of the mobile terminal, as shown in Figure 4.

The mobile terminal MN is about to leave the service range of the source access service node S-ASN and performs signal detection. It detects that the signal of the new ASN is greater than the signal of the S-ASN, and the difference in signal strength exceeds a certain threshold. The MN sets this ASN to The destination access service node T-ASN triggers network layer switching.

Step S303-1: MN triggers network switching and sends a session modification request to AME, and the request message contains the session ID;

Step S303-2: AME receives the session modification request and sends the session modification request to the session management network element SME corresponding to the session according to the session ID. After receiving the request, SME feeds back a session modification response to AME to prepare for the session modification;

Step S303-3: After receiving the response, the AME sends a session modification command to the MN, and the MN starts the timer and performs network switching;

Step S303-4: The MN initiates a routing request to the T-ASN. The request contains authorization information and requests access to the T-ASN;

Step S303-5: T-ASN receives the routing request, assigns a new routing identifier RID2 to the MN, and records the UID/RID2 mapping information into the local cache;

Step S303-6: T-ASN establishes a temporary transmission tunnel to S-ASN. At the same time, T-ASN sends a mapping information update request to ILR. After receiving the update request, ILR updates the mapping information in the local cache and records the latest MN. mapping information;

Step S303-7: T-ASN sends a successful response to the routing request to the MN and performs subsequent session modification work;

Step S303-8: After the access work is completed, the MN sends a new session establishment request to the AME through the T-ASN, and the data packet contains the IDs of the old and new sessions;

Step S303-9: After receiving the new session establishment request, the AME sends a session modification request to the SME that manages the old session, including the old and new session IDs;

Step S303-10: The SME that manages the old session determines the ASN that needs to be executed and relocates the ASN;

Step S303-11: Feed back the AME session modification response. This data packet contains the information of the new session;

Step S303-12: SME sends a session modification request to T-ASN, and T-ASN feeds back a session modification response to SME;

Step S303-13, T-ASN establishes a T-ASN-specific session with the MN, and sends a new session establishment success response to the MN after the establishment is completed;

Step S303-14: After the new session is established, the wait timer stops and the MN releases the session resources with the S-ASN.

The network layer data packet forwarding path of the session before the MN cross-gateway moves is shown as path a in Figure 1. The MN transmits data with the edge DN through S-ASN; during the MN handover process, that is, before the timer ends, the MN transmits data through the T -The temporary tunnel between ASN and S-ASN carries out data transmission with the source DN. The path is shown as path b in Figure 1; after the MN switching is completed and the APP migration is completed, the timer expires and the MN communicates with the target DN through T-ASN. Data transmission is performed directly, as shown in path c in Figure 1.

Embodiment 3: Data packet forwarding process between mobile terminal MN and WAN application platform

The mobile terminal MN not only communicates with the edge application host, but also needs to communicate with the application host in the wide area network, as shown in Figure 6. The following describes the data packet forwarding process for communication initiated by the mobile terminal.

(1) The MN sends an original data packet to the ASNm (the ASN accessed by the MN) through the RAN, requesting communication with the corresponding public network application host (hereinafter referred to as the Corresponding Node, CN). The source address in the original data packet is UIDm (the identity of the MN), and the destination address is UIDc (the identity of the CN);

(2) After receiving the original data message sent by the MN, ASNm queries the CN mapping information in the local cache according to the CN identity UIDc in the data packet, and obtains the CN routing identifier RIDc. If there is no CN mapping in the local cache information, execute (3), otherwise execute (5);

(3) ASNm sends an address query packet to ILRm in the local domain, requesting the mapping information of the CN; after receiving the query packet, ILRm decapsulates the packet to obtain UIDc, and uses DHT (Distributed Hash Table, distributed hash table) ) method to find the ILRc that stores CN mapping information;

(4) ILRc receives the query information and feeds back the CN mapping information UIDc/RIDc in the local cache to ILRm. ILRm forwards it to ASNm, and ASNm is saved in the local cache for subsequent packet forwarding;

(5) ASNm re-encapsulates the data packet, with the source address being RIDm and the destination address being RIDc, and sending it to the WAN router;

(6) The WAN router forwards the data packet to ASNc according to the destination address RIDc;

(7) After receiving the data packet, ASNc decapsulates it and obtains the original data packet destination address UIDc. ASNc sends the original data packet to the communication peer CN based on UIDc.

(8) ASNc matches the UIDm/RIDm mapping information with the mapping information in the local cache. If the match is unsuccessful, the UIDm/RIDm mapping information is stored in the local cache so that when the CN subsequently sends a data packet to the MN, the MN mapping information smooth query.

When the above-mentioned mobile terminal communicates with the public network application host, the data forwarding of the WAN and the access network is separated. Therefore, changes in the status of the terminal do not affect the network topology, and changes in the network topology will not have an impact on the terminal. It makes the terminal access method simple and does not consider the form of the terminal equipment, such as hosts, mobile devices, self-driving cars, drones, satellites, etc., and can well support users' multi-standard access.

In the above data packet forwarding process, no stateful mobility anchors are involved, which eliminates unnecessary node delays caused by mobility anchors, improves transmission reliability, and saves network resources.

In addition, the present invention also provides a deployment scheme and operation mechanism of the identity location register ILR. There are several ILRs in the network, and each ILR is responsible for storing, updating, and maintaining the latest UID/RID mapping information of the terminal. Different autonomous domains can be divided according to operators, enterprises, regions or countries. The control plane consists of several autonomous domains. All ILRs in the domain form an intra-domain routing table. The intra-domain routing table stores all terminal mapping information in the domain. At the same time, an inter-domain ILR is deployed in each autonomous domain and is responsible for communicating with the inter-domain ILRs of other domains. For data packet forwarding, all inter-domain ILRs form the inter-domain routing table.

The autonomous domain in the present invention adopts an operating mechanism based on DHT-MAP (Distributed Hash Table-Map) to store terminal mapping information and obtain mapping information of communication counterpart nodes. Each ILR can perform hash calculation on the UID to obtain the hash value. The intra-domain routing table stores the virtual coordinate area of the respective domain. The virtual coordinate area is composed of the intra-domain hash value. All hash values form a virtual coordinate space.

The following will take the communication between the mobile terminal MN and the public network application host as an example to illustrate the control plane operation mechanism.

When the MN accesses the ASNm, it sends a routing request to the ASNm. The ASNm forwards it to the AME. The AME authenticates and authorizes the MN. The ASNm allocates RIDm to the legal MN and sends the MN's mapping information UIDm/RIDm to the control plane of this domain. ILR (the ILR in this domain is not necessarily the ILRm that stores MN mapping information). Then ILR performs hash calculation on UIDm, obtains the virtual coordinate point of MN in the virtual coordinate space, and sets it as point P. If point P is in this autonomous domain, the ILR of this autonomous domain stores the mapping information in the intra-domain routing table; if it is not in this autonomous domain, the ILR forwards it to the inter-domain ILR of this autonomous domain, and the ILR sends it to the distance virtual Inter-domain ILR in an autonomous domain with closer coordinate points. This inter-domain ILR determines whether P is in the virtual coordinate area of this autonomous domain. If so, it will be stored in the intra-domain routing table; if not, it will be forwarded to a domain closer to P. Inter-ILR until P is in the target autonomous domain. The storage is successful and fed back to ASNm. ASNm sends a routing request success response to the MN.

The ILR that stores the mobile terminal MN mapping information is ILRm, and the ILR that stores the public network application host mapping information is ILRc. ILRm is deployed in autonomous domain 1, and ILRc is deployed in autonomous domain 3. The distance between the virtual coordinate area of autonomous domain 1 and the virtual coordinate area of autonomous domain 2 is closer than the distance between the virtual coordinate area of autonomous domain 1 and the virtual coordinate area of autonomous domain 3.

When the MN initiates communication to the CN, it first queries the CN's mapping information through ASNm. The local cache in ASNm does not record the CN's mapping information. Then, ASNm sends a query request packet to ILRm, and the request packet contains UIDc information. After receiving the query request packet, ILRm performs hash calculation on UIDc to obtain the hash value. Its virtual coordinate point is set to P, and it is judged whether point P is located in the virtual coordinate area of this autonomous domain. If point P is located in the virtual coordinate area, query the intra-domain routing table, obtain the CN mapping information, and feed it back to ASNm; if it is not in the virtual coordinate area, the query request is forwarded to the inter-domain ILR1, and ILR1 forwards it to the virtual domain through the inter-domain routing table. The inter-domain ILR of the autonomous domain closer to point P in the coordinate space, that is, ILR2. After receiving the query request packet, ILR2 determines whether point P is in this virtual coordinate area. If it is, it searches the intra-domain routing table and feeds back the CN mapping information; if it is not, it forwards it to the inter-domain ILR in the autonomous domain closer to point P. ILR3 , until point P is in the target autonomous domain, query the intra-domain routing table and feed back the CN mapping information to ILRm. Then, ILRm forwards the mapping information to ASNm, ASNm stores the mapping information in the local cache, and the mapping information query process is completed. Afterwards, the data packet is encapsulated, the source address is RIDm, the destination address is RIDc, and the data packet is sent to the WAN router. The WAN router routes the data packet to ASNc based on RIDc. Finally, ASNc sends the data packet to the corresponding host CN according to UIDc.

It should be noted that the above-mentioned embodiments are not used to limit the scope of protection of the present invention. Equivalent transformations or substitutions made on the basis of the above-mentioned technical solutions all fall within the scope of protection of the claims of the present invention.

Claims (8)

1. A relocation method to ensure MEC business continuity, characterized in that the method includes the following steps: Step S1: The mobile terminal MN goes online, completes the registration and session registration process, and binds the session to the terminal identity; Step S2: The mobile terminal MN is about to undergo cross-gateway handover, triggering the network layer and application layer notification coordination mechanism, and the application layer senses the network layer status change; In step S3, the mobile terminal MN undergoes cross-gateway handover, and collaboratively performs network layer handover management and application layer edge application migration process. 2. The relocation method to ensure MEC business continuity according to claim 1, characterized in that the step S1 is as follows, the mobile terminal MN registration and session registration process is as follows: Step S101: The mobile terminal MN goes online, has a unique user identity identifier (UID) in the entire network, sends a routing request to its access service node (ASN) according to the signal strength, and requests access to the ASN; Step S102: After receiving the routing request from the MN, the ASN sends an authentication and authorization request to the access and mobility management network element (AME) in the core network. The AME performs authentication services for the terminal requesting access and determines access. Whether the terminal is legal, AME authorizes the legal terminal; ASN allocates an appropriate routing identifier (RID) to the authorized MN, and records the MN's UID/RID mapping information into the local cache; Step S103, the ASN sends a UID/RID mapping information update request to the Identity Location Register (ILR) to inform the ILR of the latest mapping information of the mobile terminal; Step S104: After receiving the update of the mapping information, the ILR records the mapping information UID/RID of the mobile terminal into the local cache so that subsequent communication peers can initiate communication to the MN to query the routing information of the MN; Step S105: After the access work is completed, the ASN sends a successful response to the routing request to the MN. The MN client application (APP) establishes a session with the edge APP in the edge data center. This session includes multiple types, such as communication data in the 5G network. Unit (PDU) session, Packet Data Network (PDN) session in 4G network, first the MN sends a session establishment request to the AME in the core network through the ASN; Step S106: The AME selects the session management network element SME for the session that the MN needs to establish based on the MN's RID information; Step S107: The AME sends an MN session creation request to the selected session management network element (SME). After receiving the request, the SME feeds back response information to the AME; Step S108, AME and SME collaborate to authenticate the session to be established, determine whether the session to be established is legal, and authorize the legal session; Step S109: The SME sends an MN session creation request to the ASN accessed by the MN. After receiving the session request, the ASN feeds back response information to the SME; Step S110: After receiving the response information, the SME forwards the session information to be created to the AME. After receiving the response information, the AME feeds back the response information to the SME; Step S111, the ASN establishes an ASN-specific session with the mobile terminal through the access network (RAN). The session identifier is the user identity identifier UID of the MN. The establishment process includes sending session request response information to the mobile terminal MN. At this point, the session establishment between the MN and the mobile edge application (ME APP) is completed. 3. The relocation method for ensuring MEC business continuity according to claim 1, characterized in that, In step S2, the network layer and application layer notify the coordination mechanism, as follows: Step S201: The application layer subscribes to network event notifications, which is implemented through AF in three ways: Method 1: The application layer transmits the application's request for network resources through the AF request message, thereby realizing the network layer's delay and bandwidth requirements required for the service. Method 2: The application layer subscribes to event notification messages to the network layer through AF/NEE. For example, when a node switches across gateways, AF is notified immediately when the network detects that the subscribed event is triggered. Method 3: The application layer initiates the modification process of terminal data routing through AF, and adjusts the uplink and downlink and multi-homing policies to manage the access connection point ASN. Step S202: After receiving the subscription message, the AF sends an AF request affecting the routing policy control information to the PCE; Step S203: When the network layer PCE receives the AF subscription message, the network layer will send a notification message to the application layer, including the changed target data network identifier DNAI. The purpose is for the application layer to obtain the target ME host location for application migration and ensure the access node Consistency between ASN and target APP location. 4. The relocation method for ensuring MEC business continuity according to claim 1, characterized in that, The network layer switching management and application layer edge application migration process described in step S3 is as follows: The mobile terminal MN is about to leave the service range of the source access service node S-ASN and performs signal detection. It detects that the signal of the new ASN is greater than the signal of the S-ASN, and the difference in signal strength exceeds a certain threshold, and sets this ASN as the destination. Access the service node T-ASN, and the mobile application platform of T-ASN serves as the target mobile application platform T-MEP, triggering the migration of the application layer edge application ME APP. When the mobile terminal MN moves across gateways, the application layer edge application ME APP migration process is divided into 6 modules. Module 1, MN and S-ASN separation, Step S301: The MN is about to leave the service range of the S-ASN. The client APP in the MN sends an application move request to the S-MEP in the source edge network, and the S-MEP feeds back a response to the MN; Step S302: S-MEP enables application mobility to prepare for S-APP migration, which includes locating the S-APP that is in session with the MN APP and organizing session context data; Module 2, MN accesses T-ASN, Step S303: The MN network layer performs handover management and application layer session update process; Module 3, update traffic rules for MN, Step S304: The mobile terminal MN accesses the T-ASN. The S-RNIS receives the handover message of the MN and sends the MN area change notification to the S-MEP. After receiving the message, the S-MEP feeds back a confirmation response to the S-RNIS; Step S305, S-MEP queries S-MEPM for the target host T-MEC host that allows the MN to receive information. S-MEPM forwards the query to MEO. MEO passes the operator policy, mobility policy, application requirements, application capabilities, and mobility A series of strategic methods such as edge system status, MEO finally determines the T-MEC host and forwards the information to S-MEP through S-MEPM so that S-MEP can contact T-MEP. At the same time, S-MEP obtains the need to migrate in the platform S-APP and session context information to T-MEP; Step S306, S-MEP queries the affected traffic routing rules, which include the MN UID as part of the service filter and application information as part of the application program; Step S307, S-MEP sends a traffic routing rule update request to T-MEP, which contains application information and affected traffic routing rules; Step S308, the T-MEP receives the traffic routing rule update request from the S-MEP and updates the local traffic routing rules; Step S309: After the update is completed, feed back the traffic routing rule update response to the S-MEP; Module 4, Application Migration Management, Step S310: The MN network layer switching is completed, the T-MEP routing rule update is completed, and the MN APP sends an application migration request to the S-MEP; Step S311, S-MEP captures the status information of the serving S-APP and obtains the completed session context information in step S305; Step S312, S-MEP sends an application migration request to MEO through S-MEPM, including the S-APP ID and the ID of the target node; Step S313, MEO sends an instantiation application request to T-MEPM, and T-MEPM instantiates T-APP on the target T-MEC host; Step S314: The instantiation is completed, and MEO feeds back the application migration response to S-MEP through S-MEPM; Step S315, the S-MEP sends an application status information transmission request to the T-MEP. The application status information transmission request includes the captured status information and session context information of the S-APP; Step S316, T-MEP synchronizes the received service status information and context information with the T-APP running on the T-MEC host, and activates it; Step S317, T-MEP sends an application status information transmission response to S-MEP to indicate whether the service status transmitted by the application instance is successful; Step S318: After the service status information and context information are synchronized, the T-APP instance sends an application instance running notification to the T-MEP; Step S319, T-MEP forwards the T-APP application instance running notification to S-MEP; Step S320, S-MEP sends an application migration response to the MN APP to trigger the next module; Module 5, activate new traffic routing rules, Step S321, T-APP activates traffic routing rules to T-MEP; Step S322, T-MEP activates traffic routing rules; Step S323: T-MEP forwards the traffic routing rule activation notification to S-MEP; Step S324, S-APP sends a request to deactivate or update local traffic rules to S-MEP; Step S325, S-MEP receives the RNIS area change notification from S-RNIS; Step S326, when S-MEP receives the traffic routing rule activation notification from T-MEP and the traffic routing deactivation or update request from S-APP, S-MEP deactivates or updates the corresponding traffic routing rules; Module 6, terminate the source application, Step S327: If it is necessary to save resources, etc., S-MEP, S-MEPM, MEO, S-APP, and S-RNIS collaboratively terminate the source application. 5. The relocation method for ensuring MEC business continuity according to claim 4, characterized in that step S303 network layer switching management process and application session update process are as follows: Step S303-1: MN triggers network switching and sends a session modification request to AME, and the request message contains the session ID; Step S303-2: AME receives the session modification request and sends the session modification request to the session management network element SME corresponding to the session according to the session ID. After receiving the request, SME feeds back a session modification response to AME to prepare for the session modification; Step S303-3: After receiving the response, the AME sends a session modification command to the MN, and the MN starts the timer and performs network switching; Step S303-4: The MN initiates a routing request to the T-ASN. The request contains authorization information and requests access to the T-ASN; Step S303-5: T-ASN receives the routing request, assigns a new routing identifier RID2 to the MN, and records the UID/RID2 mapping information into the local cache; Step S303-6: T-ASN establishes a temporary transmission tunnel to S-ASN. At the same time, T-ASN sends a mapping information update request to ILR. After receiving the update request, ILR updates the mapping information in the local cache and records the latest MN. mapping information; Step S303-7: T-ASN sends a successful response to the routing request to the MN and performs subsequent session modification work; Step S303-8: After the access work is completed, the MN sends a new session establishment request to the AME through the T-ASN, and the data packet contains the IDs of the old and new sessions; Step S303-9: After receiving the new session establishment request, the AME sends a session modification request to the SME that manages the old session, including the old and new session IDs; Step S303-10: The SME that manages the old session determines the ASN that needs to be executed and relocates the ASN; Step S303-11: Feed back the AME session modification response. This data packet contains the information of the new session; Step S303-12: SME sends a session modification request to T-ASN, and T-ASN feeds back a session modification response to SME; Step S303-13, T-ASN establishes a T-ASN-specific session with the MN, and sends a new session establishment success response to the MN after the establishment is completed; Step S303-14: After the new session is established, the wait timer stops and the MN releases the session resources with the S-ASN. 6. The structure to implement the relocation method for ensuring MEC business continuity according to any one of claims 1 to 5, characterized in that the structure includes a network layer structure and an application layer structure, The network layer structure is divided into central DC (Data Center, i.e. data center), local DC, and edge DC. Application resources are offloaded to locations closer to users. Mobile terminals include multiple types. 7. The structure of the relocation method to ensure MEC business continuity according to claim 6, characterized in that the network layer structure includes Access service node ASN, ASN is the interface between the access network and the core network, and is also the access connection point of the mobile terminal MN. It is responsible for ensuring the communication connection between the MN and the entire network, assigning the location address RID to the MN, registration, authentication, and authorization. , Traffic accounting, caching the UID/RID mapping relationship of the mobile terminal at the control level. In the data level, ASN is responsible for encapsulating and decapsulating data packets in the network. Identity Location Register (Identifier Locator Register), ILR is responsible for maintaining and managing the latest UID/RID mapping relationship of the mobile terminal, the initial registration process, and the RID query of the WAN communication peer. Session management network element SME, SME is responsible for session management, tunnel maintenance, route identification allocation and management, ASN selection, policy implementation, QoS control and traffic accounting, Access and mobility management network element AME. AME is the access point of terminals and wireless core networks. It is responsible for performing registration, connection, reachability, and mobility management, and provides session management information transmission channels for mobile terminals and SMEs. Terminal access provides authentication and authentication functions. Policy Control Network Element PCE (Policy Control Network Element), PCE supports a unified policy framework for controlling network behavior, provides policy rules for control plane execution, and accesses subscription information related to policy formulation in the database. Network capability exposure network element NEE (Network Exposure Network Element), NEE provides external exposure of network functional capabilities. External exposure can be divided into monitoring capabilities, supply capabilities, traffic routing applications and impacts, and policies/planning. spending ability. 8. The structure of the relocation method to ensure MEC business continuity according to claim 6, characterized in that the application layer structure includes: Mobile edge application ME APP. ME APP is an application instance running on the virtualized infrastructure. It interacts with the mobile edge platform to obtain the service-oriented open capabilities of the mobile edge platform. Mobile edge platform MEP, MEP is responsible for providing mobile edge services for APP, including: service registration, service discovery, status monitoring, local offloading, DNS service, load balancer, firewall, as well as wireless network information service, location information service, and bandwidth management service and a series of wireless network capability services; Radio Network Information Service RNIS (Radio Network Information Service), RNIS provides wireless network related information services for ME APP. ME APP can obtain the desired wireless network information by consulting or subscribing. Mobile edge platform manager MEPM. MEPM is responsible for the basic operation and maintenance of MEP, mobile edge service configuration, life cycle management of ME APP, and application rules and demand management of ME APP. Mobile edge orchestrator MEO, MEO is the orchestration center of MEC services, located in the central DC. MEO macroscopically controls all the resources and capacities of the MEC platform, mainly including computing, storage, network resources, application mirroring resources in the virtual infrastructure manager, Check the integrity and authenticity of the software package, and then measure the user resource requirements and the available resources of each ME host to select the most appropriate ME host for deployment. Application Function AF (Application Function), AF refers to various services in the application layer of the core network. Data forwarding plane software DP (Data Plane), DP is responsible for executing the traffic rules issued by the MEP and processing traffic between ME APP, ME service, DNS server, proxy server, 3GPP network, other access networks, LAN and external networks.