CN-122001802-A - Route optimization method, route optimization device, electronic equipment and storage medium

Abstract

The invention provides a route optimization method, a device, electronic equipment and a storage medium, wherein the method comprises the steps of responding to a route request, obtaining network state data of a target network and map data of a geographic area where the target network is located; the method comprises the steps of carrying out feature extraction on a route request to obtain route request feature data, carrying out feature extraction on network state data to obtain network state feature data, carrying out feature extraction on map data to obtain map feature data, inputting the route request feature data, the network state feature data and the map feature data into a route optimization model to obtain an optimal route, wherein the route optimization model is obtained by training based on sample route request feature data, sample network state feature data and sample map feature data and an optimal route label. According to the method, the optimal route path is predicted through the route optimization model, the multidimensional characteristic data is comprehensively considered, and the instantaneity and the accuracy of route optimization are improved.

Inventors

• XIE YANNA
• SUN XIAOLIE

Assignees

• 浪潮通信信息系统有限公司

Dates

Publication Date

20260508

Application Date

20251224

Claims (10)

1. 1. A method of route optimization, comprising: responding to a routing request, and acquiring network state data of a target network and map data of a geographic area where the target network is located; Extracting features of the routing request to obtain routing request feature data, extracting features of the network state data to obtain network state feature data, and extracting features of the map data to obtain map feature data; And inputting the route request feature data, the network state feature data and the map feature data into a pre-constructed route optimization model to obtain an optimal route path output by the route optimization model, wherein the route optimization model is obtained by training based on the sample route request feature data, the sample network state feature data and the sample map feature data and an optimal route path label.
2. 2. The route optimization method according to claim 1, wherein the route optimization model comprises a feature fusion layer, a graph annotation layer and a path decision layer, wherein the feature fusion layer is used for fusing the route request feature data, the network state feature data and the map feature data to obtain fusion feature vectors of each route node of the target network, the graph annotation layer is used for calculating association relations among a plurality of route nodes based on the fusion feature vectors of each route node, and the path decision layer is used for calculating the optimal route path based on the fusion feature vectors of each route node and the association relations among a plurality of route nodes.
3. 3. The route optimization method according to claim 2, wherein the calculating the association relationship between the plurality of routing nodes based on the fusion feature vector of each of the routing nodes comprises: Determining the attention weight between any two routing nodes in a plurality of routing nodes based on the fusion feature vector of each routing node; and calculating the association relation among a plurality of routing nodes based on the fusion feature vector of each routing node and the attention weight among any two routing nodes in the plurality of routing nodes.
4. 4. The route optimization method according to claim 1, wherein the route request comprises source node information, destination node information, constraint conditions and optimization targets, wherein the optimization targets comprise path performance optimization and/or path cost optimization, and wherein the constraint conditions comprise performance constraint conditions and/or cost constraint conditions.
5. 5. The route optimization method of claim 1, wherein the network state data comprises performance state data and location state data for each routing node of the target network, the performance state data comprising load state data and latency state data.
6. 6. The route optimization method according to claim 1, wherein after obtaining the optimal route path output by the route optimization model, further comprising: And performing visual rendering on the optimal routing path based on the map data.
7. 7. The route optimization method according to claim 1, wherein the route optimization model is trained based on the steps of: Determining a sample routing request, acquiring sample network state data of a sample network, and acquiring sample map data of a geographic area where the sample network is located; Extracting features of the sample routing request to obtain sample routing request feature data, extracting features of the sample network state data to obtain sample network state feature data, and extracting features of the sample map data to obtain sample map feature data; Determining an optimal routing path label corresponding to the sample routing request; And training an initial route optimization model by taking the sample route request feature data, the sample network state feature data and the sample map feature data as sample data and taking an optimal route path label corresponding to the sample route request as a sample label, and obtaining the route optimization model after training is completed.
8. 8. A route optimization device, comprising: The system comprises an acquisition unit, a routing request acquisition unit and a control unit, wherein the acquisition unit is used for responding to the routing request and acquiring network state data of a target network and map data of a geographic area where the target network is positioned; The feature extraction unit is used for carrying out feature extraction on the routing request to obtain routing request feature data, carrying out feature extraction on the network state data to obtain network state feature data, and carrying out feature extraction on the map data to obtain map feature data; The prediction unit is used for inputting the route request feature data, the network state feature data and the map feature data into a pre-built route optimization model to obtain an optimal route path output by the route optimization model, wherein the route optimization model is obtained by training based on sample route request feature data, sample network state feature data and sample map feature data and an optimal route path label.
9. 9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, characterized in that the processor implements the route optimization method according to any of claims 1 to 7 when executing the computer program.
10. 10. A non-transitory computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the route optimization method according to any one of claims 1 to 7.

Description

Route optimization method, route optimization device, electronic equipment and storage medium Technical Field The present invention relates to the field of communications technologies, and in particular, to a route optimization method, a route optimization device, an electronic device, and a storage medium. Background In the field of network communication, in particular to a 4G/5G mobile communication network, the planning and optimization of end-to-end routing are core links for guaranteeing the service quality and the stable operation of the network. The routing in the prior art relies mostly on conventional routing protocols based on network topology or simple dynamic index calculations. The method generally abstracts the network into a set of logic nodes and links, and makes path decisions mainly according to parameters of network layers such as hop count, bandwidth and the like, so that real-time performance and accuracy are poor, and a global optimal path is difficult to generate in a complex and changeable network environment. Disclosure of Invention The invention provides a route optimization method, a route optimization device, electronic equipment and a storage medium, which are used for solving the defect that the route optimization method in the prior art is poor in instantaneity and accuracy. The invention provides a route optimization method, which comprises the following steps: responding to a routing request, and acquiring network state data of a target network and map data of a geographic area where the target network is located; Extracting features of the routing request to obtain routing request feature data, extracting features of the network state data to obtain network state feature data, and extracting features of the map data to obtain map feature data; And inputting the route request feature data, the network state feature data and the map feature data into a pre-constructed route optimization model to obtain an optimal route path output by the route optimization model, wherein the route optimization model is obtained by training based on the sample route request feature data, the sample network state feature data and the sample map feature data and an optimal route path label. In some embodiments, the route optimization model includes a feature fusion layer, a graph meaning layer and a path decision layer, wherein the feature fusion layer is used for fusing the route request feature data, the network state feature data and the map feature data to obtain a fusion feature vector of each route node of the target network, the graph meaning layer is used for calculating an association relationship among a plurality of route nodes based on the fusion feature vector of each route node, and the path decision layer is used for calculating the optimal route path based on the fusion feature vector of each route node and the association relationship among a plurality of route nodes. In some embodiments, the calculating the association relationship between the plurality of routing nodes based on the fused feature vector of each routing node includes: Determining the attention weight between any two routing nodes in a plurality of routing nodes based on the fusion feature vector of each routing node; and calculating the association relation among a plurality of routing nodes based on the fusion feature vector of each routing node and the attention weight among any two routing nodes in the plurality of routing nodes. In some embodiments, the routing request includes source node information, destination node information, constraints, and optimization objectives, including path performance optimization and/or path cost optimization, and constraints, including performance constraints and/or cost constraints. In some embodiments, the network state data includes performance state data and location state data for each routing node of the target network, the performance state data including load state data and latency state data. In some embodiments, after obtaining the optimal routing path output by the routing optimization model, the method further includes: And performing visual rendering on the optimal routing path based on the map data. In some embodiments, the route optimization model is trained based on the following steps: Determining a sample routing request, acquiring sample network state data of a sample network, and acquiring sample map data of a geographic area where the sample network is located; Extracting features of the sample routing request to obtain sample routing request feature data, extracting features of the sample network state data to obtain sample network state feature data, and extracting features of the sample map data to obtain sample map feature data; Determining an optimal routing path label corresponding to the sample routing request; And training an initial route optimization model by taking the sample route request feature data, the sample network state feature data an