Fiber Optic Data Service Expansion Through ML Area Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deploying, upgrading, and extending communication networks is costly, and existing methods lack efficient strategies for optimizing fiber optic data service expansion based on network inventory, development infrastructure, and user information.
Innovation Solution
A prediction model using machine learning is employed to generate precedence metrics for new fiber optic data service deployment, considering network inventory data, development infrastructure information, and user information, enabling targeted network reconfiguration actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fiber optic data service is deployed in all geographic areas, then user coverage is improved, but deployment cost increases
Solution Approach 1:
The patent applies local quality by differentiating deployment strategies across different geographic areas based on predicted precedence metrics. Instead of uniform deployment, the system identifies high-value areas with higher precedence metrics and prioritizes deployment there, while deferring or skipping low-value areas. This localized differentiation optimizes the balance between coverage and cost.
Solution Approach 2:
The patent utilizes parameter changes by employing machine learning models to predict precedence metrics based on various input parameters such as network inventory data, development infrastructure information, and user information. These predicted metrics serve as dynamic parameters that guide deployment decisions, allowing the system to adapt deployment priorities based on changing conditions and predictions.
2Measurement precision
If network inventory data and development infrastructure information are collected for all areas, then deployment accuracy is improved, but data collection cost increases
Solution Approach 1:
The patent applies preliminary action by collecting and preparing network inventory data, development infrastructure information, and user information before the actual deployment decision-making process. This pre-collection of data enables the machine learning models to generate accurate precedence metrics in advance, improving deployment accuracy without requiring data collection during the deployment process itself.
Solution Approach 2:
The system employs self-service by using machine learning models that automatically process collected data and generate deployment predictions without requiring manual analysis. The models self-process the network inventory, infrastructure, and user data to produce precedence metrics, reducing the need for human intervention in data processing and thereby controlling data collection costs while maintaining accuracy.
3Measurement precision
If machine learning prediction models are trained extensively, then precedence metric accuracy is improved, but model training time and computational cost increase
Solution Approach 1:
The patent applies partial action by training machine learning models on a representative subset of the available data rather than requiring complete data processing. The models are trained to generate sufficient accuracy for deployment decisions without processing every possible data point or using excessive computational resources. This balanced approach achieves adequate precision while controlling training time and computational costs.
Data Source
AI summary
A processing system may obtain network inventory data, development infrastructure information, and user information to train a prediction model to predict precedence metrics for a new fiber optic data service in geographic areas that do not include a current fiber optic data service of the communication network, the prediction model comprising at least one machine learning model. The processing system may then generate a precedence metric of a geographic area that does not include the current fiber optic data service by applying to the prediction model: current network inventory data, current development infrastructure information, and current user information, where an output of the prediction model comprises the at least one precedence metric. In addition, the processing system may perform a network reconfiguration action in the communication network in response to the at least one precedence metric of the geographic area.