IP management method of UPF pool

Abstract

The invention provides an IP management method of a UPF pool, which comprises the following steps: firstly, subdividing an address pool according to a certain management granularity to obtain routing environment information, associating a corresponding and matched address field for the UPF in a user IP address pool by an SMF according to the environment information, and establishing a coupling relationship between the address field and the UPF; when a user session request is processed on the SMF, an anchor point UPF is selected according to the session request and user information, an IP is allocated to a user from an address field associated with the UPF, tunnel information establishment to the UPF is initiated, and a subsequent session processing flow is carried out; at the moment, the core network is the IP address specified by the user session and the anchored N6 port, namely, the correct routing can be realized. According to the invention, the user flow backhaul routing during networking in the pool group mode can be effectively realized, the pool group mode networking, cross-DC networking and session management of the core network are more practical and manageable, and the defect that the IP is allocated by the UPF is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of 5G core network, and specifically to an IP management method of a UPF pool. Background Art

[0002] In the large-scale deployment of 5G networks, the core network 5G is usually deployed based on the cloud platform, and resource network elements such as SMF and UPF will be networked in a pool group to improve the reliability of the core network and increase the service processing capacity. This involves unified resource management of the network element collection that constitutes the pool group. For example, in session management, SMF needs to perform unified IP address management on the pooled UPF under its control, as well as some specific user IP traffic routing requirements generated by this, especially when deployed across data centers.

[0003] Problems and shortcomings of existing technologies: In the session management of the 5G core network, IP address allocation can be performed on SMF or UPF. In practice, the control plane session management network element SMF generally allocates IP addresses to service users, and notifies the UPF that carries the session of the IP address through the N4 session to establish a context for user data forwarding management. When networking in the UPF pool group mode, SMF needs to establish N4 association links with multiple UPFs and allocate IP addresses to users on these UPFs. One problem facing UPF pool group networking is how to manage the respective backhaul routes of users on each UPF.

[0004] Normally, each UPF instance has an independent N6 port address as the downstream entry for user data packets. If the same SMF address pool is used to allocate IPs to users distributed on these UPFs, when the downstream data packets of these users reach the bearer network of the UPF, the switch or router of the data bearer network will not be able to decide which UPF's N6 port to deliver the data packet to. Usually, the routers or switches of the bearer network perform routing / forwarding processing based on the subnet (address segment). At this time, it will not be possible to perform subnet-based routing on these UPFs on the switch, and routing for each user IP will bring too many rules and resource usage on the switch of the bearer network, which is impractical. If we consider that the UPF address pool will be deployed across data centers, users may be distributed across DCs, which will make the backhaul of user data more complicated.

[0005] One solution is to transfer address allocation to UPF, but this will have the following problems: network maintainers need to make unified static plans, IP address resources are bound to each UPF, and it is inconvenient to uniformly manage the core network. Summary of the invention

[0006] In order to solve the problems of the prior art, the present invention provides an IP management method for a UPF pool, which can perceive the UPF location and routing environment, and can effectively implement the user traffic return routing when the pool group is networked. It is suitable for both the UPF pool within a single data center and the UPF pool group across DCs, thereby making the core network's pool group networking, cross-DC networking and session management more practical and manageable, and also avoiding the defects of IP allocation by UPF.

[0007] For the pool-based networking of UPF, the solution first subdivides the address pool according to a certain management granularity, and divides a large address pool into schedulable address segments according to routable subnets. When a UPF instance is started and enters service, when registering its own instance information with NRF, in addition to carrying the conventional UPF Nf Profile information, UPF also carries its N6 port address and the routing environment information of the bearer network of the N6 port, such as which IP subnets are planned and configured for return routes. When SMF discovers UPF through NRF and establishes an N4 association link with SMF, it obtains this routing environment information. When it establishes an N4 association link Association with UPF, SMF associates the corresponding and matching address segments for UPF in the user IP address pool based on this environment information, and establishes a coupling relationship between these address segments and UPF.

[0008] When processing a user session request on the SMF, it will first select the anchor UPF based on the session request and user information, that is, the UPF with the N6 port, then allocate an IP to the user from the address segment associated with the UPF, and then initiate the establishment of tunnel information to the UPF, and perform subsequent session processing procedures.

[0009] At this time, the core network specifies the IP address for the user session and the anchored N6 port, which can achieve correct routing.

[0010] The specific process is as follows: 1. The bearer network equipment connected to UPF, such as switches, configure the current UPF user subnet to point to the UPF N6 address according to the user address planning and overall network design; 2. In the startup parameters of UPF, configure the subnet information of the connected bearer network device. After the initial startup of UPF, when registering with NRF, in addition to the standard protocol information, it carries these subnets, N6 port address and other routing environment information.

[0011] 3. NRF registers UPF information, such as UPF instance ID, supported slices, etc., and sets it to the in-service state. When there are network elements that subscribe to UPF information, NRF initiates a notification message to these network elements (usually SMF network elements) 4. The SMF network element receives the NRF notification or actively discovers the NRF at regular intervals, and receives the NRF response. In the response, the NRF returns the additional routing environment information to the SMF. Based on this information, the SMF determines the appropriate address segment (matching the received subnet information) from the address pool and assigns it to the UPF. During the life cycle of the UPF, this address segment will be bound until the N4 node association between the UPF and the SMF is disconnected.

[0012] 5. SMF and UPF establish an N4 association link, and UPF can process SMF's command request; when SMF receives a session request, it selects UPF to this UPF, calls the address segment of this UPF to allocate an IP address to the user, and performs subsequent processing.

[0013] The beneficial effects of the present invention are: 1. It can perceive the UPF location and routing environment, and can effectively implement the user traffic backhaul routing when this pool group networking is used. It is applicable to both the UPF pool within a single data center and the UPF pool group across DCs, making the core network's pool group networking, cross-DC networking and session management more practical and manageable, while also avoiding the defects of IP allocation by UPF.

[0014] 2. When the 5G core network is networked in the UPF pool group, SMF can uniformly schedule the user IP address resources of the UPF pool, especially in the UPF pool composed of cross-data centers, avoiding the drawbacks of using UPF to allocate user IP addresses. This centralized control of IP addresses can greatly improve the efficiency of network operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of dynamic association of UPF address; Figure 2 This is a specific implementation case diagram. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] For the pool-based networking of UPF, the solution first subdivides the address pool according to a certain management granularity, and divides a large address pool into schedulable address segments according to routable subnets. When a UPF instance is started and enters service, when registering its own instance information with NRF, in addition to carrying the conventional UPF Nf Profile information, UPF also carries its N6 port address and the routing environment information of the bearer network of the N6 port, such as which IP subnets are planned and configured for return routes. When SMF discovers UPF through NRF and establishes an N4 association link with SMF, it obtains this routing environment information. When it establishes an N4 association link Association with UPF, SMF associates the corresponding and matching address segments for UPF in the user IP address pool based on this environment information, and establishes a coupling relationship between these address segments and UPF.

[0019] When processing a user session request on the SMF, it will first select the anchor UPF based on the session request and user information, that is, the UPF with the N6 port, then allocate an IP to the user from the address segment associated with the UPF, and then initiate the establishment of tunnel information to the UPF, and perform subsequent session processing procedures.

[0020] At this time, the core network specifies the IP address for the user session and the anchored N6 port, which can achieve correct routing.

[0021] The specific process is as follows: 1. The bearer network equipment connected to UPF, such as switches, configure the current UPF user subnet to point to the UPF N6 address according to the user address planning and overall network design; 2. In the startup parameters of UPF, configure the subnet information of the connected bearer network device. After the initial startup of UPF, when registering with NRF, in addition to the standard protocol information, it carries these subnets, N6 port address and other routing environment information.

[0022] 3. NRF registers UPF information, such as UPF instance ID, supported slices, etc., and sets it to the in-service state. When there are network elements that subscribe to UPF information, NRF initiates a notification message to these network elements (usually SMF network elements) 4. The SMF network element receives the NRF notification or actively discovers the NRF at regular intervals, and receives the NRF response. In the response, the NRF returns the additional routing environment information to the SMF. Based on this information, the SMF determines the appropriate address segment (matching the received subnet information) from the address pool and assigns it to the UPF. During the life cycle of the UPF, this address segment will be bound until the N4 node association between the UPF and the SMF is disconnected.

[0023] 5. SMF and UPF establish N4 association link, UPF can process SMF's instruction request; when SMF receives the session request, it selects UPF to this UPF, calls the address segment of this UPF to allocate IP address to the user, and performs subsequent processing. Figure 1 shown.

[0024] The following is a specific implementation example of this method. Figure 2 shown.

[0025] UPF1 and UPF2 in the UPF pool group managed by SMF belong to two geographically different data centers (DCs) DC1 and DC2. The slices, DNN and other information of the two services are consistent. The N6 port address of UPF1 is 10.10.16.6, and the N6 port address of UPF2 is 20.10.16.6.

[0026] In the traditional user address management mode, SMF uses a unified address allocation for all UPFs under its jurisdiction, and the user addresses of UPFs in all pool groups come from a large address pool of SMF, such as the 192.168.0.0 / 255.255.0.0 subnet. When using this traditional method to manage UPFs in the above network topology, SMF will indiscriminately allocate IPs from this address pool to users under all UPFs, which results in addresses from subnet 192.168.0.0 / 16 being randomly distributed on UPFs of DC1 and DC2, but the N6 port address of each UPF is different. When a downlink data packet needs to be sent to the UE, the bearer network switch / router cannot determine which UPF's N6 port to send the data to based on the subnet of the user IP.

[0027] When the method of the present invention is used, SMF will split a large address pool 192.168.0.0 / 16 into smaller address segments, 192.168.1.0 / 24, 192.168.5.0, 24, etc., and assign these address segments to UPF when establishing N4 link association with SMF after UPF goes online based on the bearer network routing environment information connected to UPF N6 port. If the information of 192.168.1.0, 24 subnet and UPF1 has been configured on the bearer network switch of UPF1, SMF will assign the address segment 192.168.1.0 to UPF1 based on this environment information, and similarly assign the address segment 192.168.5.0 to UPF2 based on the environment information of UPF2.

[0028] When a user initiates a PDU session request, the SMF first selects the UPF (UPF with N6 egress) as the anchor point for the user based on the user's slice, DNN, location and other information, and allocates an IP address to the user in the address segment managed by this UPF, and establishes the resources required for the session in this UPF. For example, the SMF selects UPF1 and allocates 192.168.1.21 to the UE. Since the bearer network has established a route with this address segment and the N6 port address, no matter which UPF in the UPF pool the UE chooses to anchor the session, the correct return of the user traffic can be achieved.

[0029] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, the above is only a preferred embodiment of the present invention. Since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited to this. Any technician familiar with the technical field is within the technical scope disclosed by the present invention. For ordinary technicians in the technical field, changes or replacements that can be easily thought of should be covered within the protection scope of the present invention without departing from the principle of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A UPF pool IP management method, characterized in that: First, the address pool is subdivided according to a certain management granularity, and a large address pool is divided into schedulable address segments according to routable subnets. When SMF discovers UPF through NRF and establishes an N4 association link with SMF, it obtains these routing environment information. When it establishes an N4 association link Association with UPF, SMF associates the corresponding and matching address segments for UPF in the user IP address pool based on these environment information, and establishes a coupling relationship between these address segments and UPF; when processing user session requests on SMF, first select the anchor UPF according to the session request and user information, that is, the UPF with the N6 port, and then allocate an IP to the user from the address segment associated with the UPF, and then initiate the establishment of tunnel information to UPF, and perform subsequent session processing procedures; at this time, the core network specifies the IP address for the user session and the anchored N6 port, and can achieve correct routing.

2. The IP management method of the UPF pool according to claim 1, characterized in that: The specific process is as follows: 1) The bearer network equipment connected to the UPF, such as a switch, configures the current UPF user subnet to point to the UPF N6 address according to the user address planning and overall network design; 2) In the UPF startup parameters, configure the subnet information of the connected bearer network device; After the UPF is initially started, when registering with the NRF, it carries standard protocol information and routing environment information; 3) NRF registers UPF information, such as UPF instance ID, supported slices, etc., and sets it to the in-service state; when there are network elements that subscribe to UPF information, NRF initiates notification messages to these network elements; 4) The SMF network element receives the NRF notification or actively discovers the NRF at regular intervals, and receives the NRF response. In the response, the NRF returns the additional routing environment information to the SMF; based on this information, the SMF determines the address segment that matches the received subnet information from the address pool and assigns it to the UPF. During the life cycle of the UPF, this address segment will be bound until the N4 node association between the UPF and the SMF is disconnected; 5) SMF and UPF establish an N4 association link, and UPF processes the command request of SMF; when SMF receives a session request, it selects UPF to this UPF, calls the address segment of this UPF to allocate an IP address to the user, and performs subsequent processing; at this time, the core network specifies the IP address for the user session and anchors the N6 port, which can achieve correct routing.

3. The IP management method of the UPF pool according to claim 2, characterized in that: Step 2) The routing environment information includes subnet information and N6 port address information.

4. The IP management method of the UPF pool according to claim 1, characterized in that: Step 3) The UPF information includes the UPF instance ID and supported slices.

5. The IP management method of the UPF pool according to claim 1, characterized in that: Step 3) The network element that subscribes to UPF information is an SMF network element.

Images