System and method for generating topology template and cloud premise integrated topology

TWI946205BActive Publication Date: 2026-11-11CHUNGHWA TELECOM CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
TW114147248
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-12-03
Publication Date
2026-11-11
Estimated Expiration
2045-12-02

Smart Images

  • Figure TWG2TB001920957_001
    Figure TWG2TB001920957_001
  • Figure TWG2TB001920957_002
    Figure TWG2TB001920957_002
  • Figure TWG2TB001920957_003
    Figure TWG2TB001920957_003
Patent Text Reader

Abstract

A system and method for generating topology templates and cloud-ground integrated topologies are provided. The method includes the following steps: a ground module transmits topology structure data corresponding to a topology graph to a cloud module; an AI topology template generation module uses the topology structure data and a graph neural network to obtain topology groups, and the AI ​​topology template generation module uses the topology groups to generate a topology template; and a cloud-ground integrated topology module uses the topology structure data to generate a cloud-ground integrated topology.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A system for generating topology templates and cloud-ground integrated topologies, comprising: Ground module; The system includes a cloud module and a ground module, which are communicatively connected to the ground module. The cloud module includes an AI topology template generation module and a cloud-ground integrated topology module. The AI ​​topology template generation module includes an AI clustering unit and a topology template generation unit. The ground module transmits topology structure data corresponding to the topology graph to the cloud module. The AI ​​topology template generation module uses the topology structure data and a graph neural network (GNN) to obtain topology groups, and uses the topology groups to generate a topology template. The AI ​​clustering unit uses the graph neural network and feature vectors to obtain graph-level embedding vectors, and performs a clustering operation on the graph-level embedding vectors to obtain the topology groups. The topology template generation unit uses the graph-level embedding vectors and the topology groups to obtain group-level embedding vectors, and uses the group-level embedding vectors to generate the topology template. The cloud-ground integrated topology module uses the topology structure data to generate a cloud-ground integrated topology.

2. The system as described in claim 1, wherein the ground module includes topology template resources, network device resources, and a topology generation module, wherein the ground module selects a pre-stored topology template from the topology template resources according to user operation, and the ground module selects a resource list corresponding to the pre-stored topology template from the network device resources; the topology generation module generates the topology map according to the pre-stored topology template, the resource list, and the user operation.

3. The system as described in claim 2, wherein the cloud module includes a topology storage module and a topology resource, wherein the ground module uses the topology map to obtain the topology data; the ground module transmits the topology data to the cloud module; and the topology resource stores the topology data.

4. The system as described in claim 1, wherein the AI ​​topology template generation module includes a data cleaning unit and a feature extraction unit, wherein the data cleaning unit performs a data cleaning operation on the topology data to obtain the cleaned topology data; and the feature extraction unit uses the cleaned topology data to convert the feature vector.

5. The system as described in claim 1, wherein the cloud module includes a topology template delivery module, wherein the ground module includes a template storage module and topology template resources, wherein the topology template delivery module transmits the topology template to the ground module; and the template storage module stores the topology template in the topology template resources.

6. The system as described in claim 1, wherein the cloud module includes topology resources, ground-side mapping resources, and cross-border circuit resources, wherein the topology resources store the topology data, wherein the ground-side mapping resources store the ground-side IP, ground-side name, and ground-side identity verification credentials corresponding to the ground-side module, wherein the cross-border circuit resources store the ground-side usage circuits corresponding to the ground-side module, wherein the cloud-ground integrated topology module uses a cloud central node, the topology data, the ground-side IP, the ground-side name, the ground-side identity verification credentials, and the ground-side usage circuits to generate the cloud-ground integrated topology.

7. The system as described in claim 6, wherein the local IP address, the local name, and the local authentication credentials correspond to a standardized API interface, wherein the cloud-to-ground integration topology module calls the standardized API interface to obtain statistical data corresponding to the local module; and the cloud-to-ground integration topology module uses the statistical data to update the cloud-to-ground integration topology.

8. The system as described in claim 1, wherein the ground module is located on the enterprise intranet and the cloud module is located in the telecommunications operator's data center.

9. A method for generating a topology template and a cloud-ground integrated topology, suitable for a system including a ground module and a cloud module, wherein the cloud module includes an AI topology template generation module and a cloud-ground integrated topology module, wherein the AI ​​topology template generation module includes an AI clustering unit and a topology template generation unit, wherein the method includes the following steps: the ground module transmits topology structure data corresponding to a topology graph to the cloud module; the AI ​​topology template generation module uses the topology structure data and a graph neural network (GNN) to obtain topology groups, and the AI ​​topology template generation module uses the topology groups to generate a topology template, wherein the AI ​​clustering unit uses the graph neural network and feature vectors to obtain graph-level embedding vectors, and the AI ​​clustering unit performs a clustering operation on the graph-level embedding vectors to obtain the topology groups; the topology template generation unit uses the graph-level embedding vectors and the topology groups to obtain group-level embedding vectors, and the topology template generation unit uses the group-level embedding vectors to generate the topology template; and the cloud-ground integrated topology module uses the topology structure data to generate a cloud-ground integrated topology.

Citation Information

Patent Citations

  • Topological graph construction method and device, equipment, storage medium and product

    CN118827394A

  • Method and system for topological operation

    TW201929502A

  • Network topology model generation and deployment for machine learning systems

    US20230032585A1