VoLTE Diagnostic Engine Using ML for Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Diagnosing the root causes of Quality of Experience (QoE) problems in VoLTE calls is challenging due to complex network architecture and multi-layer dependencies, making it difficult for telecommunication carriers to identify and address issues such as high call drop rates and audio quality issues effectively using traditional methods.
Innovation Solution
A machine learning-based approach is employed using performance monitoring software on user devices to collect performance indicators, which are then analyzed by a VoLTE diagnostic engine to extract critical features and apply decision trees and classifiers to determine the root causes of QoE problems, generating responses for network operators and device manufacturers to resolve these issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional diagnostic methods are used to identify QoE problems in VoLTE calls, then the network architecture complexity and multi-layer dependencies make it difficult to determine root causes, but implementing machine learning-based diagnosis increases system complexity and data processing requirements
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the complex network data and the diagnostic conclusion. The model processes multiple performance indicators, call features, and network parameters through trained algorithms to identify root causes, acting as a mediator that translates complex multi-layer data into actionable diagnostic insights without requiring manual analysis of the entire system complexity
Solution Approach 2:
The patent applies preliminary action by pre-training machine learning models with historical call data and performance indicators before deployment. The models are prepared in advance with knowledge of various QoE problem patterns, enabling them to quickly diagnose root causes during actual operations without needing to analyze the full complexity of network architecture in real-time
2Productivity
If manual user feedback collection methods are used, then the feedback process is delayed due to manual input time and suffers from insufficient sample size, but automated performance monitoring increases data collection infrastructure complexity
Solution Approach 1:
The patent implements self-service by having user devices automatically collect and report their own performance indicators without requiring manual user intervention. The devices self-monitor call setup time, audio quality metrics, and network performance parameters, automatically generating feedback data that is transmitted to the diagnostic system, thereby eliminating manual input delays and reducing the burden on users
Solution Approach 2:
The patent establishes a continuous feedback loop where performance indicators are collected from user devices, processed through machine learning models, and used to generate diagnostic conclusions that are fed back to network operators. This automated feedback mechanism replaces manual rating scales with systematic data collection and analysis, improving both efficiency and sample size while maintaining manageable infrastructure through standardized data formats
3Measurement precision
If comprehensive performance indicators are collected from multiple device layers, then diagnostic accuracy improves, but data processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-processing and filtering performance indicators before they reach the diagnostic model. Relevant features are identified and extracted in advance from the comprehensive data set, reducing the dimensionality of input data. This preliminary feature engineering maintains diagnostic accuracy by preserving critical information while reducing the computational burden of processing all raw data
Solution Approach 2:
The patent segments the diagnostic process into distinct stages: data collection from multiple device layers, feature extraction and selection, model inference, and result generation. By dividing the comprehensive data processing into manageable segments, the system can process performance indicators efficiently at each stage without being overwhelmed by the total data volume, thereby reducing overall processing time while maintaining diagnostic precision
Data Source
AI summary
A VoLTE diagnostic engine may receive VoLTE call records of VoLTE calls that are carried by a wireless carrier network for multiple user devices. The VoLTE call records may include performance indicators and call features for the voice calls. Each call feature of a VoLTE call may represent a circumstance under which the VoLTE call is established and ended. The VoLTE diagnostic engine may apply a decision tree to the VoLTE call records to identify critical features of one or more call conditions that lead to Quality of Experience (QoE) problems for the VoLTE calls captured in the VoLTE call records. Each call condition may include a subset of the call features. Further, the VoLTE diagnostic engine may apply a classifier on the critical features to determine a root cause of a QoE problem for at least one call conditions.