CN-121985369-A - Wireless network fault positioning method, device, electronic equipment and storage medium

CN121985369ACN 121985369 ACN121985369 ACN 121985369ACN-121985369-A

Abstract

The invention provides a wireless network fault positioning method, a device, electronic equipment and a storage medium, wherein the method comprises the steps of acquiring equipment attribute data, topological structure data and alarm time sequence data of a wireless network; the method comprises the steps of processing equipment attribute data, topological structure data and alarm time sequence data to obtain embedded vectors of all nodes of a wireless network, inputting the embedded vectors of all the nodes into a graph neural network model to obtain a fault root positioning result of the wireless network output by the graph neural network model, wherein the graph neural network model is obtained by training based on embedded vector samples of all node samples of a wireless network sample and fault root positioning result labels of the wireless network sample. According to the invention, the normalized node embedding vector is obtained by fusing the multi-source heterogeneous data of the wireless network, the graph neural network model is adopted, the fault root cause of the wireless network is positioned based on the node embedding vector, and the efficiency and the accuracy of fault positioning are improved.

Inventors

YU JIE
WANG WEI

Assignees

浪潮通信信息系统有限公司

Dates

Publication Date: 20260505
Application Date: 20260122

Claims (10)

1. A method for locating a wireless network failure, comprising: acquiring equipment attribute data, topology structure data and alarm time sequence data of a wireless network; processing the equipment attribute data, the topological structure data and the alarm time sequence data to obtain embedded vectors of all nodes of the wireless network; The embedded vector of each node is input into a graph neural network model, and a fault root positioning result of the wireless network output by the graph neural network model is obtained, wherein the graph neural network model is obtained by training based on embedded vector samples of each node sample of a wireless network sample and a fault root positioning result label of the wireless network sample.
2. The method for locating a fault in a wireless network according to claim 1, wherein the processing the device attribute data, the topology data, and the alert timing data to obtain the embedded vector of each node of the wireless network comprises: inputting the equipment attribute data, the topological structure data and the alarm time sequence data into a Embedding model to obtain embedded vectors of the nodes output by the Embedding model; The Embedding model is obtained by training based on a device attribute data sample, a topological structure data sample and an alarm time sequence data sample of a wireless network sample and embedded vector labels of all node samples of the wireless network sample.
3. The wireless network fault location method of claim 2, wherein the Embedding model comprises: The attribute embedding layer is used for encoding the equipment attribute data to obtain attribute embedding vectors of the nodes; The structure embedding layer is used for encoding the topological structure data to obtain structure embedding vectors of the nodes; the alarm embedding layer is used for encoding the alarm time sequence data to obtain alarm embedding vectors of the nodes; and the space-time fusion layer is used for fusing the attribute embedded vector, the structure embedded vector and the alarm embedded vector of each node based on the attention mechanism to obtain the embedded vector of each node.
4. The wireless network fault location method of claim 1, wherein the graph neural network model comprises: the graph attention layer is used for dynamically calculating the attention weight between each node and each corresponding neighbor node, updating the embedded vector of each node based on the attention weight, and obtaining the feature vector of each node; and the hole time sequence convolution layer is used for positioning the fault root cause of the wireless network based on the characteristic vector of each node to obtain the fault root cause positioning result.
5. The method for positioning a fault root cause of a wireless network according to claim 1, wherein after obtaining the positioning result of the fault root cause of the wireless network output by the graph neural network model, the method further comprises: Visually displaying the fault root cause positioning result; and sending the fault root cause positioning result to the client.
6. The method for locating a fault in a wireless network according to claim 2, wherein after obtaining the location result of the fault root cause of the wireless network output by the graph neural network model, the method further comprises: after performing fault processing on the wireless network based on the fault root cause positioning result, acquiring fault processing feedback information; determining a first performance evaluation result of the Embedding model and determining a second performance evaluation result of the graph neural network model; updating parameters of the Embedding model based on the fault handling feedback information and the first performance evaluation result, and updating parameters of the graph neural network model based on the fault handling feedback information and the second performance evaluation result.
7. The wireless network fault location method of claim 1, wherein the graph neural network model is trained based on the steps of: acquiring a device attribute data sample, a topological structure data sample and an alarm time sequence data sample of a wireless network sample, and determining a fault root cause positioning result label of the wireless network sample; Processing the equipment attribute data sample, the topological structure data sample and the alarm time sequence data sample to obtain embedded vector samples of each node sample of the wireless network sample; The embedded vector samples of the node samples are input into an initial graph neural network model, and a predicted fault root cause positioning result of the wireless network samples output by the initial graph neural network model is obtained; And calculating a loss function value based on the predicted fault root cause positioning result and the fault root cause positioning result label, and performing iterative optimization on parameters of the initial graph neural network model based on the loss function value to obtain the graph neural network model.
8. A wireless network fault location device, comprising: the acquisition unit is used for acquiring equipment attribute data, topological structure data and alarm time sequence data of the wireless network; The processing unit is used for processing the equipment attribute data, the topological structure data and the alarm time sequence data to obtain embedded vectors of all nodes of the wireless network; The fault positioning unit is used for inputting the embedded vector of each node into a graph neural network model to obtain a fault root cause positioning result of the wireless network output by the graph neural network model, wherein the graph neural network model is obtained by training based on embedded vector samples of each node sample of a wireless network sample and a fault root cause positioning result label of the wireless network sample.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the wireless network fault location method of any of claims 1 to 7 when the computer program is executed by the processor.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the wireless network fault location method according to any of claims 1 to 7.

Description

Wireless network fault positioning method, device, electronic equipment and storage medium Technical Field The present invention relates to the field of network operation and maintenance technologies, and in particular, to a method and apparatus for locating a wireless network fault, an electronic device, and a storage medium. Background With the rapid development of wireless communication technology, the size and complexity of wireless networks are increasing. In a huge wireless network, the occurrence of faults is unavoidable, and how to quickly and accurately locate the faults of the wireless network is a key problem to be solved. The traditional wireless network fault positioning method has a plurality of limitations that on one hand, the traditional method is mainly based on manual experience and preset rules to carry out simple statistical analysis, so that complex network data are difficult to deeply excavate and effectively characterize, and on the other hand, the traditional method only considers single characteristics of a wireless network, so that the fault positioning accuracy is low, the efficiency is low, and the current wireless network operation and maintenance requirements cannot be met. Disclosure of Invention The invention provides a wireless network fault positioning method, a wireless network fault positioning device, electronic equipment and a storage medium, which are used for solving the defects of low efficiency and poor accuracy of the wireless network fault positioning method in the prior art. The invention provides a wireless network fault positioning method, which comprises the following steps: acquiring equipment attribute data, topology structure data and alarm time sequence data of a wireless network; processing the equipment attribute data, the topological structure data and the alarm time sequence data to obtain embedded vectors of all nodes of the wireless network; The embedded vector of each node is input into a graph neural network model, and a fault root positioning result of the wireless network output by the graph neural network model is obtained, wherein the graph neural network model is obtained by training based on embedded vector samples of each node sample of a wireless network sample and a fault root positioning result label of the wireless network sample. In some embodiments, the processing the device attribute data, the topology data, and the alert timing data to obtain an embedded vector of each node of the wireless network includes: inputting the equipment attribute data, the topological structure data and the alarm time sequence data into a Embedding model to obtain embedded vectors of the nodes output by the Embedding model; The Embedding model is obtained by training based on a device attribute data sample, a topological structure data sample and an alarm time sequence data sample of a wireless network sample and embedded vector labels of all node samples of the wireless network sample. In some embodiments, the Embedding model includes: The attribute embedding layer is used for encoding the equipment attribute data to obtain attribute embedding vectors of the nodes; The structure embedding layer is used for encoding the topological structure data to obtain structure embedding vectors of the nodes; the alarm embedding layer is used for encoding the alarm time sequence data to obtain alarm embedding vectors of the nodes; and the space-time fusion layer is used for fusing the attribute embedded vector, the structure embedded vector and the alarm embedded vector of each node based on the attention mechanism to obtain the embedded vector of each node. In some embodiments, the graph neural network model includes: the graph attention layer is used for dynamically calculating the attention weight between each node and each corresponding neighbor node, updating the embedded vector of each node based on the attention weight, and obtaining the feature vector of each node; and the hole time sequence convolution layer is used for positioning the fault root cause of the wireless network based on the characteristic vector of each node to obtain the fault root cause positioning result. In some embodiments, after the obtaining the fault root cause positioning result of the wireless network output by the graph neural network model, the method further includes: Visually displaying the fault root cause positioning result; and sending the fault root cause positioning result to the client. In some embodiments, after the obtaining the fault root cause positioning result of the wireless network output by the graph neural network model, the method further includes: after performing fault processing on the wireless network based on the fault root cause positioning result, acquiring fault processing feedback information; determining a first performance evaluation result of the Embedding model and determining a second performance evaluation result of the graph neural network model;