Optical Link Fault Identification Using Dynamic Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optical fiber communication systems face challenges with low fault handling efficiency and long rectification delays due to ineffective alert prompts and inaccurate fault mode identification, leading to complex operations and user impact.
Innovation Solution
An optical link fault identification method that extracts feature parameters from performance data, such as receive optical power, to accurately identify fault modes without manual threshold settings, using a fault mode identification model trained with historical data for timely and efficient fault processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional alert system with manually specified baseline thresholds is used, then the system can detect faults, but the accuracy is low due to missed alerts or false alerts
Solution Approach 1:
The patent transforms fixed manually-specified thresholds into dynamic thresholds that adapt based on historical performance data and environmental conditions. The system continuously learns from past data to adjust detection parameters, thereby improving both accuracy and reliability of fault detection without manual intervention.
Solution Approach 2:
The system performs self-diagnosis and self-adjustment by automatically analyzing its own performance data to establish dynamic thresholds. This eliminates dependence on manual threshold configuration and enables the system to autonomously improve its detection accuracy over time through continuous learning from operational data.
2Productivity
If manual troubleshooting is performed based on alert prompts, then faults can be identified, but the efficiency is low and the delay in fault rectification is excessive
Solution Approach 1:
The patent replaces manual mechanical troubleshooting with an automated intelligent diagnosis system that uses machine learning algorithms to analyze performance data and identify fault modes. This substitution eliminates the time-consuming manual analysis process and enables rapid automated fault identification and rectification recommendations.
Solution Approach 2:
The system performs preliminary fault analysis and generates diagnostic results automatically as soon as anomalies are detected, eliminating the need to wait for manual intervention. By pre-configuring automated diagnosis workflows and having the system continuously monitor and analyze data, fault rectification can begin immediately without human delay.
3Device complexity
If performance data at a single time point is used for determination, then the process is simple, but missed alerts or false alerts are prone to occur
Solution Approach 1:
The patent implements periodic sampling and continuous monitoring of performance data over time, analyzing trends across multiple time points rather than relying on single instantaneous measurements. This temporal dimension adds robustness to fault detection by distinguishing genuine faults from transient anomalies, improving precision without requiring overly complex real-time analysis.
Solution Approach 2:
The system performs preliminary analysis on historical performance data to establish baseline behaviors and normal variation patterns before actual fault detection occurs. This pre-processing creates a reference framework that simplifies real-time monitoring while significantly improving detection precision by comparing current data against learned historical patterns.
Data Source
AI summary
This application provides an optical link fault identification method, and relate to the field of communications technologies. The method includes: obtaining performance data of a network device, extracting a feature parameter of the performance data, and identifying a fault mode on an optical link based on the feature parameter. The method resolves problems of a difficulty in fault identification and slow troubleshooting that are caused by a large quantity of devices, many line faults, and a difficulty in obtaining manual troubleshooting cases. In addition, a fault can be quickly identified when the fault occurs, improving troubleshooting efficiency. When an optical link risk does not cause a fault, deterioration of the performance data can be found in advance based on a feature, to perform identification and warning.