Cell- and UE-Level Machine Learning for Cellular Troubleshooting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cellular network troubleshooting methods struggle to efficiently identify and resolve user-level performance degradation issues due to their reliance on cell-level detection and limited ground truth data, leading to prolonged resolution times and manual investigation efforts.
Innovation Solution
A data-driven approach utilizing deep neural networks to analyze complex spatial-temporal features from network logs, incorporating cell-level and UE-level machine learning models to automatically distinguish between network and device-related issues, reducing manual intervention and improving resolution efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual troubleshooting methods are used to diagnose service issues at per-UE level, then diagnostic accuracy can be achieved through advanced expertise and in-depth analysis, but the cost and time consumption increase significantly due to the need for substantial amounts of network log data analysis
Solution Approach 1:
The patent creates a digital twin or virtual representation of the network environment that replicates real network conditions and behaviors. This virtual model allows operators to analyze and diagnose issues without directly interacting with the actual network, thereby reducing troubleshooting time while maintaining diagnostic accuracy through the copied network state
Solution Approach 2:
The system performs preliminary analysis and preparation by pre-processing network log data and creating ready-to-analyze virtual network states before actual troubleshooting occurs. This preliminary action reduces the time needed during actual diagnostic operations while maintaining accuracy through pre-computed network representations
2Measurement precision
If manual troubleshooting methods are used to diagnose service issues at per-UE level, then diagnostic accuracy can be achieved through advanced expertise and in-depth analysis, but the cost increases due to the need for advanced expertise and substantial amounts of network log data analysis
Solution Approach 1:
By creating a virtual copy of the network environment, the system consolidates complex analysis capabilities into a single virtual platform. This reduces the need for multiple expert analysts and complex manual processes, thereby reducing cost while maintaining diagnostic accuracy through the virtualized analysis environment
Solution Approach 2:
The patent replaces manual mechanical troubleshooting processes with automated virtualized systems. Machine learning algorithms and automated analysis tools substitute for human experts, reducing the cost associated with advanced expertise while maintaining or improving diagnostic accuracy through consistent automated analysis
3Productivity
If automated troubleshooting systems are implemented to reduce manual intervention, then troubleshooting efficiency improves, but the ability to handle complex user-level performance degradation issues deteriorates due to lack of advanced expertise
Solution Approach 1:
The virtual network environment replicates complex network conditions and behaviors, enabling automated systems to practice and refine their diagnostic capabilities in a realistic virtual setting. This allows automated troubleshooting to maintain high accuracy for complex issues while improving efficiency through automation
Solution Approach 2:
The system dynamically adjusts analysis parameters and data processing depth based on the complexity of the detected issue. For complex user-level performance degradation, the system automatically increases analysis depth and applies more sophisticated algorithms, maintaining accuracy while keeping the system automated and efficient
Data Source
AI summary
Aspects of the subject disclosure may include, for example, training a cell-level machine learning model to predict a likelihood of a cell site in a cellular network having service issues that impact customers of the cellular network, training a user equipment (UE) level machine learning model using output information from the cell-level machine learning model and historical information about UE-level performance metrics, receiving, from a customer associated with a UE device operating on the cellular network, information about a service degradation experienced by the customer on the UE device, providing the information about the service degradation to the UE-level machine learning model; and receiving, from the UE-level machine learning model, information identifying a source of the service degradation. Other embodiments are disclosed.