Optical network fault detection method and system

Abstract

The invention provides an optical network fault detection method and system. The method comprises the following steps: obtaining to-be-detected optical network performance data; and inputting to-be-detected optical network performance data into a trained optical network fault detection model, obtaining a hidden space reconstruction error of the to-be-detected optical network performance data, and obtaining an optical network fault detection result according to the hidden space reconstruction error and a preset fault threshold, wherein the trained optical network fault detection model is obtained by training an improved auto-encoder neural network according to sample optical network performance data, and the improved auto-encoder neural network comprises a first encoder, a second encoder and a decoder. According to the method, the optical network fault detection result is obtained according to the hidden space reconstruction error and the preset fault threshold, the anti-interference performance of the model is improved, and based on the improved auto-encoder neural network, the data mode of training data can be better learned, the reconstruction error is reduced, and the anomaly detection effect is improved.

Description

technical field [0001] The invention relates to the technical field of communications, in particular to a method and system for detecting an optical network fault. Background technique [0002] Most of the modern communications are based on the optical network. Once the optical network fails, it will cause immeasurable losses, so the fault detection in the optical network is very important. [0003] Optical network hardware faults are mainly divided into two categories, dominant faults and recessive faults. Dominant faults refer to those faults caused by a single feature obviously exceeding the normal range. A simple machine learning method or even a single-threshold discrimination method can obtain high discrimination accuracy and F1 value for dominant faults. Hidden faults refer to the faults where the equipment has obviously failed, but all the eigenvalues ​​are within the normal range. It is difficult to obtain good detection results for hidden faults through simple mac...

AI Technical Summary

Problems solved by technology

Unbalanced data has a great impact on the training of the model. It will make the model more inclined to judge the data as normal in exchange for a smaller loss value, resulting in poor final detection results.

Images