Article: Evaluation of hidden danger types of optical channel performance degradation based on machine learning cascading technology Journal: International Journal of Embedded Systems (IJES) 2023 Vol.16 No.3 pp.205 - 214 Abstract: In order to improve the accuracy of performance degradation hazard type assessment and improve the transmission effect of optical channel signal, this paper uses machine learning cascade technology to conduct in-depth research on optical channel performance degradation hazard type assessment. This paper firstly analyses the causes and characteristics of potential degradation hazards of optical channel (OC) performance, and then classifies OC performance degradation hazards by using machine learning algorithm. Meanwhile, in order to verify the effectiveness of the machine learning cascade technology, this paper takes the real OC performance degradation data of an optical communication enterprise as a sample set to conduct precision experiment analysis. The results show that the ML algorithm can effectively and accurately classify the potential degradation hazards of OC performance. By concatenating decision tree, support vector machine and neural network, the accuracy of identifying potential degradation hazards of OC performance can reach 92.37%. Inderscience Publishers - linking academia, business and industry through research

Title: Evaluation of hidden danger types of optical channel performance degradation based on machine learning cascading technology

Authors: Qing Wang; Qiong Cheng; Yuzhi Jing; Shuxin Nie; Jun Yang

Addresses: State Grid Yangquan Power Supply Company, Yangquan, Shanxi, China ' State Grid Yangquan Power Supply Company, Yangquan, Shanxi, China ' State Grid Yangquan Power Supply Company, Yangquan, Shanxi, China ' State Grid Yangquan Power Supply Company, Yangquan, Shanxi, China ' State Grid Yangquan Power Supply Company, Yangquan, Shanxi, China

Abstract: In order to improve the accuracy of performance degradation hazard type assessment and improve the transmission effect of optical channel signal, this paper uses machine learning cascade technology to conduct in-depth research on optical channel performance degradation hazard type assessment. This paper firstly analyses the causes and characteristics of potential degradation hazards of optical channel (OC) performance, and then classifies OC performance degradation hazards by using machine learning algorithm. Meanwhile, in order to verify the effectiveness of the machine learning cascade technology, this paper takes the real OC performance degradation data of an optical communication enterprise as a sample set to conduct precision experiment analysis. The results show that the ML algorithm can effectively and accurately classify the potential degradation hazards of OC performance. By concatenating decision tree, support vector machine and neural network, the accuracy of identifying potential degradation hazards of OC performance can reach 92.37%.

Keywords: optical channel; machine learning; types of potential performance degradation hazards; cascade technology; support vector machine; SVM.

DOI: 10.1504/IJES.2023.139055

International Journal of Embedded Systems, 2023 Vol.16 No.3, pp.205 - 214

Received: 20 Apr 2023
Accepted: 23 Jan 2024
Published online: 10 Jun 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Evaluation of hidden danger types of optical channel performance degradation based on machine learning cascading technology

Keep up-to-date