Optical Fiber Health Analysis Using Historical Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional optical networks lack comprehensive monitoring of both short-term and long-term metrics for fiber spans and connections, leading to missed potential issues, particularly within intra-NE connections, which can impact signal quality and are more accessible for remediation, and existing proactive approaches fail to provide actionable reasoning for predicted failures.

Innovation Solution

A system that logs and analyzes both short-term and long-term data from optical networks, using supervised Machine Learning techniques and expert rules to classify issues, providing detailed reports on an interactive UI to help network operators understand and prioritize fiber maintenance and routing decisions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional reactive monitoring with hard-coded thresholds is used, then alarm detection is simple and direct, but potential issues are not identified until failures occur and historical trends are not analyzed

Engineering Contradiction:

Improvefiber connection reliabilityVSAvoidhistorical performance data utilization

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system performs preliminary actions by continuously collecting and storing historical performance data (insertion loss, return loss, OTDR traces) before failures occur. This enables proactive identification of degrading connections through trend analysis, allowing maintenance to be scheduled before actual failures impact network reliability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by comparing current performance metrics against historical baselines and detecting deviations that indicate degrading connections. This continuous feedback loop enables the system to identify potential issues early and alert operators before hard-coded thresholds are breached, improving reliability while utilizing historical information.

Inventive Principle:

Principle #23Feedback

2Reliability

If ML algorithms are used for predicting failures, then proactive detection is enabled, but the algorithms act as black boxes without providing reasoning for predictions

Engineering Contradiction:

Improvefailure prediction capabilityVSAvoidreasoning transparency

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system uses physics-based models and expert rules as intermediaries between raw sensor data and failure predictions. These interpretable models analyze historical trends, environmental factors, and connection parameters to generate predictions with clear reasoning, allowing operators to understand why a connection is predicted to fail while maintaining proactive detection capability.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Ease of manufacture

If only external line fibers are monitored, then power monitoring and OTDR readings are sufficient, but intra-NE connections are overlooked despite being more vulnerable and accessible

Engineering Contradiction:

Improvemonitoring implementation simplicityVSAvoidcomprehensive fiber health monitoring

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The system implements multi-functionality by creating a unified monitoring framework that handles both external line fibers and intra-NE connections through the same technical approach. The system collects insertion loss, return loss, and OTDR data for all fiber types and applies consistent analysis methods, enabling comprehensive monitoring without significantly increasing implementation complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Loss of information

If comprehensive historical data collection and analysis is implemented, then fiber health understanding is improved, but system complexity and data processing requirements increase

Engineering Contradiction:

Improveperformance data availabilityVSAvoiddata analysis system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The system extracts only the most relevant features from comprehensive historical data, such as insertion loss trends, return loss variations, and OTDR trace changes. By focusing on key performance indicators rather than processing all raw data, the system maintains improved fiber health understanding while reducing computational complexity and data processing requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS12149284B2Analyzing performance of fibers and fiber connections using long-term historical data

Publication Date: 2024.11.19 CIENA CORP

AI summary

Systems, methods, and computer-readable media are provided for logging long-term data and analyzing the long-term data with short-term data to determine the health of fiber connections in an optical network. A method, according to one implementation, includes a step of obtaining data associated with performance of fiber connections of an optical network. The fiber connections include at least an inter-node fiber connecting two adjacent network nodes and an intra-node fiber connection connecting two photonic devices within each of the two adjacent network nodes. The method further includes the step of logging the data over time as historical data and then analyzing the health of the fiber connections based on the historical data and newly-obtained data. Also, the method includes displaying a report on an interactive user interface, whereby the report is configured to show the health of the fiber connections.