Routing optimization method and device, electronic equipment and storage medium
By acquiring network status and map data in 4G/5G mobile communication networks, and utilizing feature fusion and graph attention layer to calculate the optimal routing path, the real-time and accuracy issues of routing selection in existing technologies are solved, achieving more efficient path optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- INSPUR TIANYUAN COMM INFORMATION SYST CO LTD
- Filing Date
- 2025-12-24
- Publication Date
- 2026-05-08
AI Technical Summary
Existing routing methods suffer from poor real-time performance and accuracy in 4G/5G mobile communication networks, making it difficult to generate globally optimal paths in complex and ever-changing network environments.
By acquiring network status data and geographic area map data, feature extraction is performed and the data is input into the routing optimization model. The optimal routing path is calculated using a feature fusion layer, a graph attention layer, and a path decision layer. The model is trained based on sample data.
It improves the real-time performance and accuracy of route optimization, generating better path decisions.
Smart Images
Figure CN122001802A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communication technology, and in particular to a routing optimization method, apparatus, electronic device, and storage medium. Background Technology
[0002] In the field of network communication, especially in 4G / 5G mobile communication networks, end-to-end routing planning and optimization are core aspects of ensuring service quality and stable network operation. Existing routing technologies largely rely on traditional routing protocols based on network topology or simple dynamic metric calculations. These methods typically abstract the network as a collection of logical nodes and links, primarily making path decisions based on network-level parameters such as hop count and bandwidth. This results in poor real-time performance and accuracy, making it difficult to generate globally optimal paths in complex and ever-changing network environments. Summary of the Invention
[0003] This invention provides a routing optimization method, apparatus, electronic device, and storage medium to address the shortcomings of existing routing optimization methods in terms of poor real-time performance and accuracy.
[0004] This invention provides a routing optimization method, comprising: In response to a routing request, obtain network status data of the target network and map data of the geographical area where the target network is located; Feature extraction is performed on the routing request to obtain routing request feature data; feature extraction is performed on the network status data to obtain network status feature data; and feature extraction is performed on the map data to obtain map feature data. The routing request feature data, the network status feature data, and the map feature data are input into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model. The routing optimization model is trained based on sample routing request feature data, sample network status feature data, sample map feature data, and optimal routing path labels.
[0005] In some embodiments, the routing optimization model includes a feature fusion layer, a graph attention layer, and a path decision layer; the feature fusion layer is used to fuse the routing request feature data, the network state feature data, and the map feature data to obtain a fused feature vector for each routing node in the target network; the graph attention layer is used to calculate the association relationship between multiple routing nodes based on the fused feature vector of each routing node; and the path decision layer is used to calculate the optimal routing path based on the fused feature vector of each routing node and the association relationship between multiple routing nodes.
[0006] In some embodiments, calculating the association relationship between multiple routing nodes based on the fused feature vector of each routing node includes: Based on the fusion feature vector of each routing node, determine the attention weight between any two routing nodes among the plurality of routing nodes; Based on the fused feature vector of each routing node and the attention weight between any two routing nodes, the association relationship between the multiple routing nodes is calculated.
[0007] In some embodiments, the routing request includes: source node information, destination node information, constraints, and optimization objectives; the optimization objectives include path performance optimization and / or path cost optimization; the constraints include performance constraints and / or cost constraints.
[0008] In some embodiments, the network status data includes performance status data and location status data for each routing node in the target network; the performance status data includes load status data and latency status data.
[0009] In some embodiments, after obtaining the optimal route path output by the route optimization model, the method further includes: Based on the map data, the optimal route path is visualized and rendered.
[0010] In some embodiments, the routing optimization model is trained based on the following steps: Determine the sample routing request, obtain the sample network status data of the sample network, and obtain the sample map data of the geographical area where the sample network is located; Feature extraction is performed on the sample routing request to obtain sample routing request feature data; feature extraction is performed on the sample network state data to obtain sample network state feature data; and feature extraction is performed on the sample map data to obtain sample map feature data. Determine the optimal route path label corresponding to the sample routing request; Using the sample route request feature data, the sample network state feature data, and the sample map feature data as sample data, and the optimal route path label corresponding to the sample route request as the sample label, an initial route optimization model is trained. After training, the route optimization model is obtained.
[0011] The present invention also provides a routing optimization device, comprising: The acquisition unit is used to acquire network status data of the target network and map data of the geographical area where the target network is located in response to a routing request. The feature extraction unit is used to extract features from the routing request to obtain routing request feature data, extract features from the network status data to obtain network status feature data, and extract features from the map data to obtain map feature data. The prediction unit is used to input the routing request feature data, the network state feature data, and the map feature data into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network state feature data, sample map feature data, and optimal routing path labels.
[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the routing optimization method as described above.
[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the routing optimization method as described above.
[0014] The routing optimization method, apparatus, electronic device, and storage medium provided by this invention, in response to a routing request, acquire network status data of the target network and map data of the geographical area where the target network is located; extract features from the routing request to obtain routing request feature data, extract features from the network status data to obtain network status feature data, and extract features from the map data to obtain map feature data; input the routing request feature data, network status feature data, and map feature data into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model, thereby improving the real-time performance and accuracy of routing optimization. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0016] Figure 1 This is a flowchart illustrating the routing optimization method provided in an embodiment of the present invention.
[0017] Figure 2 This is a flowchart illustrating the training process of the routing optimization model provided in this embodiment of the invention.
[0018] Figure 3This is a schematic diagram of the routing optimization device provided in an embodiment of the present invention.
[0019] Figure 4 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0021] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, in this invention, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.
[0022] Figure 1 This is a flowchart illustrating the routing optimization method provided in an embodiment of the present invention. Figure 1 As shown, a route optimization method is provided, including the following steps: step 110, step 120, and step 130. This method's steps are merely one possible implementation of the present invention.
[0023] Step 110: In response to the routing request, obtain the network status data of the target network and the map data of the geographical area where the target network is located.
[0024] Optionally, the routing request input by the user can be obtained directly; alternatively, if an anomaly is detected in the target network, a routing request can be automatically generated to optimize the routing path between the source node and the destination node.
[0025] The target network refers to a specific communication network; the target network may include 4G base station uplinks; the target network includes multiple routing nodes, such as base stations, switches, routers, transmission network equipment, and core network gateways.
[0026] Optionally, map data is a collection of data describing the spatial information of the geographical area where the target network is located. Map data provides the coordinates of routing nodes in the real world, as well as environmental background such as terrain and landforms. Map data can be obtained based on a Geographic Information System (GIS).
[0027] In some embodiments, a routing request includes: source node information, destination node information, constraints, and optimization objectives; the optimization objectives include path performance optimization and / or path cost optimization; the constraints include performance constraints and / or cost constraints.
[0028] Optionally, the source node information includes at least the source node's ID, location, and status; the destination node information includes at least the destination node's ID, location, and status.
[0029] Optionally, the routing request may also include path performance weights and / or path cost weights.
[0030] In some embodiments, network status data includes performance status data and location status data for each routing node in the target network; performance status data includes load status data and latency status data.
[0031] Optionally, based on source node information, destination node information, and constraints, network state data of the target network and map data of the geographical area where the target network is located can be obtained.
[0032] Optionally, in response to a routing request, real-time network status data of the target network is obtained, and map data of the geographic area where the target network is located is obtained.
[0033] Optionally, routing requests, network status data, and map data can be preprocessed.
[0034] Step 120: Extract features from routing requests to obtain routing request feature data; extract features from network status data to obtain network status feature data; and extract features from map data to obtain map feature data.
[0035] Optionally, the routing request feature data includes, but is not limited to: source node features, destination node features, constraint features, optimization target features, and service features; network status feature data includes the node status features of each routing node in the target network, and the link status features of each link in the target network, with node status features including, but not limited to: node ID, node location, and node status; map feature data includes, but is not limited to, spatial relationship features between routing nodes, environmental features, and cost features.
[0036] Step 130: Input the routing request feature data, network status feature data, and map feature data into the pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network status feature data, sample map feature data, and optimal routing path labels.
[0037] Optionally, the routing optimization model includes one or more of the following networks: deep neural networks, graph neural networks, and reinforcement learning policy networks.
[0038] Optionally, the optimal route path includes at least the order of routing nodes and the path distance; the optimal route path includes one main path and multiple alternative paths.
[0039] In some embodiments, the routing optimization model includes a feature fusion layer, a graph attention layer, and a path decision layer; the feature fusion layer is used to fuse routing request feature data, network state feature data, and map feature data to obtain a fused feature vector for each routing node in the target network; the graph attention layer is used to calculate the association relationship between multiple routing nodes based on the fused feature vector of each routing node; and the path decision layer is used to calculate the optimal routing path based on the fused feature vector of each routing node and the association relationship between multiple routing nodes.
[0040] Optionally, the feature fusion layer is used to determine the weights of the routing request feature data, network status feature data, and map feature data, and to perform weighted fusion of the routing request feature data, network status feature data, and map feature data to obtain the fused feature vector of each routing node.
[0041] In some embodiments, the association between multiple routing nodes is calculated based on the fused feature vector of each routing node, including: Based on the fusion feature vector of each routing node, determine the attention weight between any two routing nodes among multiple routing nodes; Based on the fused feature vector of each routing node and the attention weight between any two routing nodes, the association relationship between multiple routing nodes is calculated.
[0042] For example, suppose node A is a base station, and nodes B and C are its two uplink nodes. If the current routing request is a high-bandwidth video stream, and node B's link has slightly higher latency but sufficient bandwidth, while node C's link has lower latency but narrower bandwidth, then for the task of transmitting video, the model will assign a higher attention weight to B than to C at point A.
[0043] In this embodiment of the invention, in response to a routing request, network status data of the target network and map data of the geographical area where the target network is located are obtained; features are extracted from the routing request to obtain routing request feature data, features are extracted from the network status data to obtain network status feature data, and features are extracted from the map data to obtain map feature data; the routing request feature data, network status feature data, and map feature data are input into the routing optimization model to obtain the optimal routing path output by the routing optimization model, thereby improving the real-time performance and accuracy of routing optimization.
[0044] In some embodiments, after obtaining the optimal route path output by the route optimization model, the method further includes: Based on map data, the optimal route path is visualized and rendered.
[0045] Optionally, different colors can be used to distinguish between the primary path and alternative paths.
[0046] Figure 2 This is a flowchart illustrating the training process of the routing optimization model provided in an embodiment of the present invention. Figure 2 As shown, in some embodiments, the routing optimization model is trained based on the following steps: Step 210: Determine the sample routing request, obtain the sample network status data of the sample network, and obtain the sample map data of the geographical area where the sample network is located; Step 220: Extract features from the sample routing request to obtain sample routing request feature data; extract features from the sample network state data to obtain sample network state feature data; extract features from the sample map data to obtain sample map feature data. Step 230: Determine the optimal route path label corresponding to the sample routing request; Step 240: Using sample route request feature data, sample network state feature data, and sample map feature data as sample data, and the optimal route path label corresponding to the sample route request as the sample label, train the initial route optimization model. After training, the route optimization model is obtained.
[0047] Optionally, the initial route optimization model includes an initial feature fusion layer, an initial graph attention layer, and an initial path decision layer. The initial feature fusion layer is used to fuse sample route request feature data, sample network state feature data, and sample map feature data to obtain a fused feature vector for each sample route node in the sample network. The initial graph attention layer is used to calculate the correlation between multiple sample route nodes based on the fused feature vector of each sample route node. The initial path decision layer is used to calculate the optimal predicted route path based on the fused feature vector of each initial route node and the correlation between multiple sample route nodes.
[0048] Optionally, based on the fused feature vector of each sample routing node, the association relationship between multiple sample routing nodes is calculated, including: Based on the fusion feature vector of each sample routing node, determine the attention weight between any two sample routing nodes among multiple sample routing nodes; Based on the fused feature vector of each sample routing node and the attention weight between any two sample routing nodes, the correlation between multiple sample routing nodes is calculated.
[0049] The routing optimization apparatus provided in the embodiments of the present invention is described below. The routing optimization apparatus described below and the routing optimization method described above can be referred to in correspondence.
[0050] Figure 3 This is a schematic diagram of the routing optimization device provided in an embodiment of the present invention, as shown below. Figure 3 As shown, the routing optimization device 300 includes: The acquisition unit 310 is used to acquire network status data of the target network and map data of the geographical area where the target network is located in response to a routing request. The feature extraction unit 320 is used to extract features from routing requests to obtain routing request feature data, extract features from network status data to obtain network status feature data, and extract features from map data to obtain map feature data. The prediction unit 330 is used to input routing request feature data, network status feature data and map feature data into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model. The routing optimization model is trained based on sample routing request feature data, sample network status feature data and sample map feature data, as well as the optimal routing path label.
[0051] Optionally, the routing optimization model includes a feature fusion layer, a graph attention layer, and a path decision layer; the feature fusion layer is used to fuse routing request feature data, network state feature data, and map feature data to obtain a fused feature vector for each routing node in the target network; the graph attention layer is used to calculate the association relationship between multiple routing nodes based on the fused feature vector of each routing node; and the path decision layer is used to calculate the optimal routing path based on the fused feature vector of each routing node and the association relationship between multiple routing nodes.
[0052] Optionally, based on the fused feature vector of each routing node, the association relationship between multiple routing nodes is calculated, including: Based on the fusion feature vector of each routing node, determine the attention weight between any two routing nodes among multiple routing nodes; Based on the fused feature vector of each routing node and the attention weight between any two routing nodes, the association relationship between multiple routing nodes is calculated.
[0053] Optionally, the routing request includes: source node information, destination node information, constraints, and optimization objectives; the optimization objectives include path performance optimization and / or path cost optimization; the constraints include performance constraints and / or cost constraints.
[0054] Optionally, the network status data includes performance status data and location status data for each routing node in the target network; the performance status data includes load status data and latency status data.
[0055] Optionally, the route optimization device also includes: The display module is used to visualize and render the optimal route based on map data.
[0056] Optionally, the route optimization model is trained based on the following steps: Determine the sample routing request, obtain the sample network status data of the sample network, and obtain the sample map data of the geographical area where the sample network is located; Feature extraction is performed on the sample routing requests to obtain sample routing request feature data; feature extraction is performed on the sample network state data to obtain sample network state feature data; and feature extraction is performed on the sample map data to obtain sample map feature data. Determine the optimal route path label corresponding to the sample route request; Using sample route request feature data, sample network state feature data, and sample map feature data as sample data, and the optimal route path label corresponding to the sample route request as the sample label, an initial route optimization model is trained. After training, the route optimization model is obtained.
[0057] It should be noted that the routing optimization device provided in this embodiment of the invention can implement all the method steps implemented in the above-described routing optimization method embodiment and can achieve the same technical effect. Therefore, the parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.
[0058] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 4As shown, the electronic device may include a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a routing optimization method. This method includes: in response to a routing request, acquiring network state data of the target network and map data of the geographical area where the target network is located; extracting features from the routing request to obtain routing request feature data, extracting features from the network state data to obtain network state feature data, and extracting features from the map data to obtain map feature data; inputting the routing request feature data, network state feature data, and map feature data into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network state feature data, sample map feature data, and optimal routing path labels.
[0059] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0060] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the routing optimization method provided by the above methods. The method includes: in response to a routing request, acquiring network state data of a target network and map data of the geographical area where the target network is located; extracting features from the routing request to obtain routing request feature data, extracting features from the network state data to obtain network state feature data, and extracting features from the map data to obtain map feature data; inputting the routing request feature data, network state feature data, and map feature data into a pre-constructed routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network state feature data, sample map feature data, and optimal routing path labels.
[0061] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the routing optimization method provided by the above methods. The method includes: in response to a routing request, acquiring network state data of a target network and map data of the geographical area where the target network is located; extracting features from the routing request to obtain routing request feature data, extracting features from the network state data to obtain network state feature data, and extracting features from the map data to obtain map feature data; inputting the routing request feature data, network state feature data, and map feature data into a pre-constructed routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network state feature data, sample map feature data, and optimal routing path labels.
[0062] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0063] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A route optimization method, characterized in that, include: In response to a routing request, obtain network status data of the target network and map data of the geographical area where the target network is located; Feature extraction is performed on the routing request to obtain routing request feature data; feature extraction is performed on the network status data to obtain network status feature data; and feature extraction is performed on the map data to obtain map feature data. The routing request feature data, the network status feature data, and the map feature data are input into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model. The routing optimization model is trained based on sample routing request feature data, sample network status feature data, sample map feature data, and optimal routing path labels.
2. The routing optimization method according to claim 1, characterized in that, The routing optimization model includes a feature fusion layer, a graph attention layer, and a path decision layer. The feature fusion layer fuses the routing request feature data, the network state feature data, and the map feature data to obtain a fused feature vector for each routing node in the target network. The graph attention layer calculates the association relationships between multiple routing nodes based on the fused feature vector of each routing node. The path decision layer calculates the optimal routing path based on the fused feature vector of each routing node and the association relationships between multiple routing nodes.
3. The routing optimization method according to claim 2, characterized in that, The step of calculating the association relationship between multiple routing nodes based on the fused feature vector of each routing node includes: Based on the fusion feature vector of each routing node, determine the attention weight between any two routing nodes among the plurality of routing nodes; Based on the fused feature vector of each routing node and the attention weight between any two routing nodes, the association relationship between the multiple routing nodes is calculated.
4. The routing optimization method according to claim 1, characterized in that, The routing request includes: source node information, destination node information, constraints, and optimization objectives; the optimization objectives include path performance optimization and / or path cost optimization; the constraints include performance constraints and / or cost constraints.
5. The routing optimization method according to claim 1, characterized in that, The network status data includes performance status data and location status data for each routing node in the target network; the performance status data includes load status data and latency status data.
6. The routing optimization method according to claim 1, characterized in that, After obtaining the optimal route path output by the route optimization model, the process further includes: Based on the map data, the optimal route path is visualized and rendered.
7. The routing optimization method according to claim 1, characterized in that, The routing optimization model is trained based on the following steps: Determine the sample routing request, obtain the sample network status data of the sample network, and obtain the sample map data of the geographical area where the sample network is located; Feature extraction is performed on the sample routing request to obtain sample routing request feature data; feature extraction is performed on the sample network state data to obtain sample network state feature data; and feature extraction is performed on the sample map data to obtain sample map feature data. Determine the optimal route path label corresponding to the sample routing request; Using the sample route request feature data, the sample network state feature data, and the sample map feature data as sample data, and the optimal route path label corresponding to the sample route request as the sample label, an initial route optimization model is trained. After training, the route optimization model is obtained.
8. A routing optimization device, characterized in that, include: The acquisition unit is used to acquire network status data of the target network and map data of the geographical area where the target network is located in response to a routing request. The feature extraction unit is used to extract features from the routing request to obtain routing request feature data, extract features from the network status data to obtain network status feature data, and extract features from the map data to obtain map feature data. The prediction unit is used to input the routing request feature data, the network state feature data, and the map feature data into a pre-built routing optimization model to obtain the optimal routing path output by the routing optimization model; the routing optimization model is trained based on sample routing request feature data, sample network state feature data, sample map feature data, and optimal routing path labels.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the routing optimization method as described in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the routing optimization method as described in any one of claims 1 to 7.