RAN Configuration Updates for Predicted Radio Service Degradation

Solution Overview

Problem

Current techniques for preventing service degradation in radio access networks (RANs) are reactive, manual, difficult to scale, and occur too late, leading to suboptimal user experiences and resource wastage due to service outages and disruptions.

Innovation Solution

A monitoring system that utilizes predictive machine learning models to analyze energy consumption patterns and network data flow graphs, proactively updating RAN configurations to mitigate potential service degradation by rerouting traffic and adjusting network parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual monitoring and configuration updates are used, then system complexity is reduced, but service degradation detection is too late and scalability is poor

Engineering Contradiction:

Improveautomation of service degradation detection and configuration updatesVSAvoidcomplexity of monitoring system

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by using machine learning models to predict service degradation before it actually occurs. The monitoring system analyzes energy consumption patterns and network data flow graphs proactively, identifying potential issues ahead of time and triggering configuration updates in advance, thereby preventing service degradation rather than reacting to it after occurrence.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The monitoring system implements self-service through automated configuration updates. Once service degradation is predicted, the system automatically determines and applies configuration changes to radios without requiring manual intervention. This self-service capability allows the system to maintain optimal performance autonomously, reducing the need for human operators while managing complexity through standardized automated workflows.

Inventive Principle:

Principle #25Self-service

2Reliability

If reactive monitoring is used, then system complexity is minimized, but service outages and disruptions occur leading to resource wastage

Engineering Contradiction:

Improveservice continuity and qualityVSAvoidresource wastage due to service outages

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary actions by using machine learning models to predict service degradation before it actually occurs. The monitoring system analyzes energy consumption patterns and network data flow graphs proactively, identifying potential issues ahead of time and triggering configuration updates in advance, thereby preventing service degradation rather than reacting to it after occurrence.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The monitoring system continuously collects feedback from energy consumption data and network performance metrics, feeding this information back into the machine learning models for ongoing prediction refinement. This feedback loop enables the system to adapt to changing network conditions and maintain high reliability by continuously improving its predictive accuracy and response effectiveness.

Inventive Principle:

Principle #23Feedback

3Productivity

If manual configuration updates are performed, then ease of operation is maintained, but response time is too late and scalability is limited

Engineering Contradiction:

Improvespeed of configuration update responseVSAvoidautomation of configuration updates

Core Design Contradiction:

ProductivityVSExtent of automation

Solution Approach 1:

The system performs preliminary actions by using machine learning models to predict service degradation before it actually occurs. The monitoring system analyzes energy consumption patterns and network data flow graphs proactively, identifying potential issues ahead of time and triggering configuration updates in advance, thereby preventing service degradation rather than reacting to it after occurrence.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The monitoring system implements self-service through automated configuration updates. Once service degradation is predicted, the system automatically determines and applies configuration changes to radios without requiring manual intervention. This self-service capability allows the system to maintain optimal performance autonomously, reducing the need for human operators while managing complexity through standardized automated workflows.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250374115A1Systems and methods for automatically updating radio access network configurations in case of predicted service degradation

Publication Date: 2025.12.04 VERIZON PATENT & LICENSING INC

AI summary

A device may receive topology data associated with a plurality of RANs and historical energy consumption data associated with respective radios of the plurality of RANs, and may generate feature data identifying features. The device may process the feature data to generate a trained predictive model, and may receive traffic associated with the plurality of RANs and current energy consumption data associated with the respective radios. The device may create a network data flow graph of the plurality of RANs or the respective radios based on the traffic and the current energy consumption data. The device may process the network data flow graph, with the trained predictive model, to determine energy consumption drops of the respective radios and corresponding KPI degradations of the respective radios. The device may identify a radio with an energy consumption drop above a predefined threshold, and may perform one or more actions for the radio.