ML Path Selection for Multi-Vendor ROADM Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-vendor reconfigurable optical add/drop multiplexer (ROADM) networks, conventional methods for evaluating and provisioning new wavelengths are vendor-specific and inefficient, failing to effectively predict optical performance due to proprietary software and lack of interoperability between different vendors' equipment.
Innovation Solution
The use of machine learning models to predict the optical performance of proposed paths in ROADM networks by defining feature sets that include characteristics of existing wavelengths, allowing for the deployment of new wavelengths based on predicted performance metrics, without requiring detailed knowledge of internal network equipment configurations or vendor-specific tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional vendor-specific methods are used for evaluating and provisioning wavelengths, then compatibility with specific vendor equipment is maintained, but interoperability between different vendors' equipment deteriorates
Solution Approach 1:
The patent implements a universal machine learning model that can evaluate and predict optical performance across multi-vendor ROADM networks. The system defines vendor-agnostic feature sets and uses trained ML models to provide consistent wavelength provisioning decisions regardless of the vendor equipment involved, enabling the system to serve multiple vendor ecosystems with a single unified approach
2Measurement precision
If detailed knowledge of internal network equipment configurations is used for wavelength evaluation, then prediction accuracy may be improved, but system complexity and data requirements increase
Solution Approach 1:
The patent extracts only the essential features needed for accurate optical performance prediction from the complex internal configurations of network equipment. By identifying and using a streamlined feature set that captures the most critical parameters, the system achieves reliable predictions without requiring detailed knowledge of proprietary internal equipment configurations, thus reducing complexity while maintaining accuracy
Data Source
AI summary
Devices, computer-readable media and methods are disclosed for selecting paths in reconfigurable optical add/drop multiplexer (ROADM) networks using machine learning. In one example, a method includes defining a feature set for a proposed path through a wavelength division multiplexing network, wherein the proposed path traverses at least one link in the network, and wherein the at least one link connects a pair of reconfigurable optical add/drop multiplexers, predicting an optical performance of the proposed path, wherein the predicting employs a machine learning model that takes the feature set as an input and outputs a metric that quantifies predicted optical performance, and determining whether to deploy a new wavelength on the proposed path based on the predicted optical performance of the proposed path.