Fault detection method for centralized access network
An access network and centralized technology, applied in the direction of data exchange network, digital transmission system, electrical components, etc., can solve problems such as difficulty in obtaining fault samples at one time, unsuitable for centralized access network, etc., to improve fault detection efficiency , Reduce the amount of manual participation and improve the effect of automation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0052] The method for training the first fault detector will be introduced below through an embodiment.
[0053] For most centralized access networks, the network architecture can be abstracted into multiple layers. Since each layer has different characteristics, according to an embodiment of the present invention, each layer in the network architecture of the centralized access network can be The first fault detectors are respectively trained to detect abnormal events by using the first fault detector corresponding to the layer when performing abnormal detection.
[0054] For the above embodiment, when training the first fault detector, the corresponding operating data of the layer can be selected in combination with the characteristics of each layer to be used as the training sample of the negative selection algorithm, and the same category as the training sample can be selected when implementing step 1 Anomaly detection of the data to be tested.
[0055] refer to figure 1...
Embodiment 2
[0106] According to Embodiment 2 of the present invention, a method for performing anomaly detection on a centralized access network is provided, refer to figure 2 , the method includes:
[0107] Step 1. For the centralized access network, a first fault detector is used to detect abnormal events, wherein the first fault detector is trained based on the operation data of the centralized access network itself and using a negative selection algorithm.
[0108] According to an embodiment of the present invention, the first fault detector is obtained through training using the method in Embodiment 1.
[0109] The inventors have found that the failure characteristics of the above layers conform to the following rules, that is, if the current layer fails, there is a high probability that the lower layer will also be abnormal. For example, if the indicators at the network layer are abnormal, there will usually be abnormalities at the lower network element layer. Therefore, directly...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com