Research Article
Fault Prediction Analysis of Communication Optical Fiber Based on SVM Algorithm
@INPROCEEDINGS{10.1007/978-3-030-90196-7_8, author={Pengfei Li}, title={Fault Prediction Analysis of Communication Optical Fiber Based on SVM Algorithm}, proceedings={Artificial Intelligence for Communications and Networks. Third EAI International Conference, AICON 2021, Xining, China, October 23--24, 2021, Proceedings, Part I}, proceedings_a={AICON}, year={2021}, month={11}, keywords={Optical cable fault SVM algorithm SMOTE algorithm}, doi={10.1007/978-3-030-90196-7_8} }- Pengfei Li
Year: 2021
Fault Prediction Analysis of Communication Optical Fiber Based on SVM Algorithm
AICON
Springer
DOI: 10.1007/978-3-030-90196-7_8
Abstract
Optical fiber is the basis of communication network, carrying a huge network traffic, the impact of the cable failure is significant. As a result, the fiber fault prediction is a hot research topic. In this paper, based on the basic parameters and fault information of optical fiber, Support Vector Machine (SVM) model is adopted to classify the faults. Since the cable fault is a small probability event, there is an imbalance in the distribution of the data sample. Use Synthetic Minority Oversampling Technique (SMOTE) algorithm to treat the imbalance data, and then analyze and study by using the SVM classification model.
The results show that the overall classification accuracy of the prediction model is 79.8%, and the prediction sensitivity of the fault fiber is 62.2%. The results show that the model has a good effect, and it can provide a certain auxiliary role for the operation and maintenance of communication fiber.
