Search

CN-121985429-A - Dynamic slice management method and device based on multi-level labels

CN121985429ACN 121985429 ACN121985429 ACN 121985429ACN-121985429-A

Abstract

The invention relates to a dynamic slice management method and a dynamic slice management device based on a multi-level label, which belong to the technical field of network management and comprise the steps of configuring a strategy, when a vehicle is electrified and tries to connect a network, a vehicle terminal sends a connection request message to the network, a strategy control engine detects the request message of the vehicle terminal to connect the network, determines priority labels of brand vehicle type nodes corresponding to the vehicle, group member identity priority labels and priority labels corresponding to operator service grades, determines a guarantee grade according to the three priority labels, converts the guarantee grade into a network slice identifier, sends a resource allocation instruction to a network slice manager, and dispatches traffic flow of the vehicle to a corresponding network slice example, and after receiving the resource allocation instruction, the network slice manager configures a data forwarding path through a software-defined network controller and distributes virtualized resources through a network function virtualization infrastructure manager. The invention obviously improves the utilization efficiency of network resources.

Inventors

  • Gu Zihuai

Assignees

  • 天翼物联科技有限公司

Dates

Publication Date
20260505
Application Date
20251229

Claims (10)

  1. 1. The dynamic slice management method based on the multi-level label is characterized by being applied to a system consisting of a vehicle enterprise management platform, a strategy control engine and a network slice manager, and comprising the following steps of: the vehicle enterprise management platform receives the configuration strategy operation of the management personnel and then configures the strategy; When a vehicle is electrified and tries to connect with a network, a vehicle terminal sends a connection request message to the network, a policy control engine detects the request message of the vehicle terminal for connecting with the network, and determines a priority label of a brand vehicle model node corresponding to the vehicle, a group member identity priority label and a priority label corresponding to an operator service grade according to the request message and a configuration policy; The policy control engine determines a guarantee level according to a priority label of a brand vehicle model node corresponding to the vehicle, a group member identity priority label and a priority label corresponding to an operator service level, converts the guarantee level into a network slice identifier according to a preset mapping relation, sends a resource allocation instruction to a network slice manager, and dispatches traffic flows of the vehicle to corresponding network slice examples; after receiving the resource allocation instruction, the network slice manager configures a data forwarding path through a software defined network controller, and allocates virtualized resources through a network function virtualization infrastructure manager; The system comprises a vehicle enterprise management platform, a policy control engine and a network slice manager, wherein the policy control engine in the system consisting of the vehicle enterprise management platform, the policy control engine and the network slice manager is connected with the vehicle enterprise management platform through a server interface, and the policy control engine is connected with the network slice manager through a network internal interface.
  2. 2. The multi-level tag based dynamic slice management method of claim 1, further comprising: The data acquisition module in the strategy control engine periodically acquires historical network load data, actual service quality data of each priority service, service renewal behavior of a train enterprise and service evaluation data from the unified data warehouse; The weight adjustment module in the policy control engine continuously and interactively learns the optimal weight configuration policies under different network states through the reinforcement learning model and the environment, and configures the weights of the priority labels of the brand vehicle type nodes, the weights of the group member identity labels and the weights of the priority labels corresponding to the service levels of operators according to the optimal weight configuration policies.
  3. 3. The dynamic slice management method based on the multi-level label according to claim 1, wherein the vehicle enterprise management platform receives the configuration policy operation post-configuration policy of the manager, and the method comprises the following steps: when receiving the operation of configuring the service grade and the priority label list corresponding to the service grade by the manager, the vehicle enterprise management platform interface configures the service grade and the priority label list corresponding to the service grade; When the vehicle enterprise management platform interface receives the operation of constructing the hierarchical brand vehicle model tree structure by the management personnel, constructing the hierarchical brand vehicle model tree structure; when receiving the operation that a manager defines a SIM card group of a cross-brand or a same-brand cross-vehicle type according to business requirements, a vehicle enterprise management platform interface defines the SIM card group of the cross-brand or the same-brand cross-vehicle type; And when the vehicle enterprise management platform interface receives the operation of allocating the priority labels for the terminal nodes in the tree structure and the defined SIM card groups, allocating the priority labels for the terminal nodes in the tree structure and the defined SIM card groups.
  4. 4. The dynamic slice management method based on multi-level labels according to claim 1, wherein determining the priority label of the brand model node corresponding to the vehicle, the group member identity label and the priority label corresponding to the operator service level according to the request message and the configuration policy comprises: inquiring a priority label of a brand vehicle model node corresponding to the vehicle, a priority label of group member identity and a priority label corresponding to the service grade of an operator in a configuration strategy according to the vehicle identification information and the service characteristic information in the request message; The request message carries vehicle identification information and service characteristic information.
  5. 5. The dynamic slice management method based on multi-level labels according to claim 1, wherein the policy control engine determines a security level according to a priority label of a brand model node corresponding to a vehicle, a group member identity priority label, and a priority label corresponding to an operator service level, comprising: A policy arbitration module in the policy control engine determines a guarantee level according to the priority label of the brand vehicle model node corresponding to the vehicle, the group member identity priority label and the priority label corresponding to the operator service level; the policy arbitration module comprises a three-dimensional arbitration algorithm, and the rule of the algorithm is a guarantee level=MAX.
  6. 6. The method for dynamic slice management based on multi-level labels according to claim 2, wherein the reinforcement learning model comprises a state space, an action space and a reward function; the state space comprises a network load level, the utilization rate of each slice resource and a service quality index; the action space is the adjustment operation of the weight of the priority label of the brand vehicle model node, the weight of the group member identity label and the weight of the priority label corresponding to the service level of the operator; The reward function comprehensively considers the overall throughput of the network, the service quality satisfaction of the high priority service and the service offer rate of the train enterprise.
  7. 7. The dynamic slice management method based on the multi-level label according to claim 3, wherein the service level comprises a platinum level, a gold level and a silver level, the priority label list corresponding to the service level comprises four levels of highest, high, medium and low, the content of the configuration strategy comprises the service level and the priority label list corresponding to the service level, and the priority of the terminal node in the tree structure and the priority of the SIM card group of the cross-brand or the same-brand cross-vehicle type.
  8. 8. The dynamic slice management device based on the multi-level label is characterized by being applied to a system consisting of a vehicle enterprise management platform, a strategy control engine and a network slice manager, and comprises the following components: The receiving module is used for the vehicle enterprise management platform to receive the configuration strategy operation of the management personnel and then configure the strategy; The system comprises a determining module, a policy control engine, a network management module and a network management module, wherein the determining module is used for sending a connection request message to the network by a vehicle terminal when the vehicle is electrified and tries to connect the network, detecting the request message of the vehicle terminal for connecting the network by the policy control engine, and determining a priority label of a brand vehicle model node corresponding to the vehicle, a group member identity label and a priority label corresponding to an operator service grade according to the request message and a configuration policy; The system comprises a sending module and a scheduling module, wherein the sending module and the scheduling module are used for determining a guarantee level by a policy control engine according to a priority label of a brand vehicle model node corresponding to a vehicle, a group member identity label and a priority label corresponding to an operator service level, converting the guarantee level into a network slice identifier according to a preset mapping relation, sending a resource allocation instruction to a network slice manager, and scheduling a service flow of the vehicle into a corresponding network slice instance; The configuration module is used for configuring a data forwarding path through a software defined network controller after the network slice manager receives the resource allocation instruction, and allocating virtualized resources through the network function virtualization infrastructure manager; The system comprises a vehicle enterprise management platform, a policy control engine and a network slice manager, wherein the policy control engine in the system consisting of the vehicle enterprise management platform, the policy control engine and the network slice manager is connected with the vehicle enterprise management platform through a server interface, and the policy control engine is connected with the network slice manager through a network internal interface.
  9. 9. An electronic device is characterized by comprising a processor and a memory; the processor is configured to execute a dynamic slice management method based on a multi-level tag according to any one of claims 1 to 7 by calling a program or instructions stored in the memory.
  10. 10. A computer-readable storage medium storing a program or instructions that cause a computer to execute a dynamic slice management method based on a multi-level tag according to any one of claims 1 to 7.

Description

Dynamic slice management method and device based on multi-level labels Technical Field The invention belongs to the technical field of network management, and particularly relates to a dynamic slice management method and device based on a multi-level label. Background In the field of internet of vehicles (V2X), in order to ensure communication quality, the prior art mainly relies on a network operator to preset a static policy. Operators typically configure fixed priority policies and network slicing templates in the network core network based on limited traffic types (e.g., autopilot, remote diagnostics) or broad vehicle categories. When a vehicle is accessed, the system guides the service flow to the corresponding slice according to the simple subscription data. The existing scheme has the obvious defects that firstly, the configuration is stiff and the flexibility is not available. The vehicle enterprise is used as a direct user of the service, and cannot dynamically and finely customize priority according to complex business logic (such as importance differences of different brands, vehicle types or specific test vehicle teams), so that network resources are difficult to accurately adapt to actual service demands. Secondly, the management granularity is rough. The method can not support the vehicle enterprise to implement the differentiation strategy on multiple dimensions such as a tree structure or a custom group of brand-vehicle type-vehicle system. Thirdly, the intelligent optimization capability is lacking. Policy weight is fixed, self-adjustment and optimization cannot be performed according to data such as historical network load conditions, service renewal or upgrade feedback of a vehicle enterprise, and the system operation efficiency is low. The most core technical problem is how to break the rigidifying mode of the full authority control of the operators, and design a set of dynamic resource allocation mechanism which not only allows the vehicle enterprises to carry out multidimensional and fine-grained policy customization, but also ensures the effective supervision of the operators. Disclosure of Invention In view of the shortcomings of the prior art, the invention aims to provide a dynamic slice management method and a dynamic slice management device based on a multi-level label, which can be used for partially and controllably downloading the configuration right of network resources to a vehicle enterprise, so as to realize on-demand, dynamic and refined allocation of network slice capacity, thereby remarkably improving the utilization efficiency of the network resources and the commercial value of the vehicle networking service. In a first aspect of the present invention, a dynamic slice management method based on multi-level labels is provided, The method is applied to a system consisting of a vehicle enterprise management platform, a strategy control engine and a network slice manager, and comprises the following steps: the vehicle enterprise management platform receives the configuration strategy operation of the management personnel and then configures the strategy; When a vehicle is electrified and tries to connect with a network, a vehicle terminal sends a connection request message to the network, a policy control engine detects the request message of the vehicle terminal for connecting with the network, and determines a priority label of a brand vehicle model node corresponding to the vehicle, a group member identity priority label and a priority label corresponding to an operator service grade according to the request message and a configuration policy; The policy control engine determines a guarantee level according to a priority label of a brand vehicle model node corresponding to the vehicle, a group member identity priority label and a priority label corresponding to an operator service level, converts the guarantee level into a network slice identifier according to a preset mapping relation, sends a resource allocation instruction to a network slice manager, and dispatches traffic flows of the vehicle to corresponding network slice examples; after receiving the resource allocation instruction, the network slice manager configures a data forwarding path through a software defined network controller, and allocates virtualized resources through a network function virtualization infrastructure manager; The system comprises a vehicle enterprise management platform, a policy control engine and a network slice manager, wherein the policy control engine in the system consisting of the vehicle enterprise management platform, the policy control engine and the network slice manager is connected with the vehicle enterprise management platform through a server interface, and the policy control engine is connected with the network slice manager through a network internal interface. Further, the dynamic slice management method based on the multi-level label further comprises the following