Optical Network Learning Engine for Dynamic Channel Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional optical networks operate based on passive, pre-calculated rules that do not scale well with increasing complexity, leading to sub-optimal performance and inflexible reactions to changes within the network.

Innovation Solution

An apparatus comprising a learning engine that updates a learning model using real-time network metrics to generate channel rank information, which is used by a recommendation engine to dynamically adjust channel throughput, signal paths, and spectral locations, optimizing network performance through artificial neural networks trained with span data and operational metrics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional passive pre-calculated network rules are used, then network operation is simplified, but network performance optimization and adaptability deteriorate

Engineering Contradiction:
Improvenetwork operation simplicityVSAvoidnetwork performance adaptability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent implements a feedback mechanism where network metrics are continuously collected from the optical network, processed by a learning engine to update a learning model, and used by a recommendation engine to generate optimization recommendations. This closed-loop feedback system enables the network to adapt dynamically to changing conditions while maintaining operational simplicity through automated decision-making.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The learning engine automatically updates the learning model using collected network metrics without requiring manual intervention. The system serves itself by autonomously generating optimization recommendations and implementing changes to network channel configurations, thereby maintaining simplicity while improving adaptability.

Inventive Principle:
Principle #25Self-service

2Ease of manufacture

If passive pre-calculated network rules are employed, then system verification effort is reduced, but network scalability worsens

Engineering Contradiction:
Improvesystem verification effortVSAvoidnetwork complexity scalability
Core Design Contradiction:
Ease of manufactureVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical-style pre-calculated network rules with an intelligent system based on machine learning. The learning engine automatically processes network metrics and generates optimization recommendations, eliminating the need for exhaustive manual system verification while enabling the network to scale with increasing complexity through adaptive intelligence.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If conventional capacity provisioning methods are used, then network configuration is simplified, but network performance optimization deteriorates

Engineering Contradiction:
Improvenetwork configuration simplicityVSAvoidnetwork performance optimization
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The recommendation engine dynamically changes network channel parameters such as throughput, signal path, and spectral location based on updated learning models. This allows the system to maintain simple configuration procedures while achieving continuous performance optimization through automated parameter adjustments driven by real-time network metrics.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3531580B1A method and apparatus for optimizing dynamically the operation of an optical network
Publication Date: 2021.08.11 ADVA OPTICAL NETWORKING SP ZOO
  • EP3531580B1 patent drawingFigure 1
  • EP3531580B1 patent drawingFigure 2
  • EP3531580B1 patent drawingFigure 3

AI summary

An apparatus and method for optimizing dynamically the performance of an optical network, said apparatus (1) comprising at least one learning engine (2) adapted to update a learning model in response to network metrics of said optical network (4) collected during operation of said optical network, wherein the updated learning model is used to generate channel rank information for network channels; and a recommendation engine (3) adapted to change a network channel throughput, a signal path and/or a spectral location of at least one network channel based on the channel rank information generated by the learning model of said learning engine (2).