ML Root Cause Analysis for Network Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cellular networks face significant challenges in rapidly detecting and resolving network anomalies, leading to prolonged outages and diminished user experience due to the time and human effort required for manual intervention.
Innovation Solution
The implementation of a machine learning-based root cause analysis framework that automates the discovery and diagnosis of network anomalies by processing Key Performance Indicators (KPIs) and alarm data, using techniques such as quantile regression forests and anomaly detection methods to identify issues and suggest corrective actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual intervention is used to detect and resolve network anomalies, then human expertise can be applied to complex problems, but the time required increases significantly leading to prolonged outages
Solution Approach 1:
The system enables self-service anomaly detection and resolution by automatically analyzing network data, identifying anomalies, determining root causes, and executing corrective actions without requiring continuous human intervention. The automated root cause analysis system processes network performance data, alarm information, and configuration data to independently diagnose and resolve issues.
Solution Approach 2:
The patent replaces manual mechanical analysis with automated machine learning-based root cause analysis. The system uses computational algorithms to process network data, identify patterns, and determine causes of anomalies, substituting human expert analysis with automated intelligent systems that operate continuously without fatigue.
2Productivity
If automated root cause analysis is implemented, then the speed of anomaly detection improves, but the system complexity increases
Solution Approach 1:
The system segments the complex anomaly detection process into distinct functional modules: data collection module, anomaly detection module, root cause analysis module, and remediation execution module. Each module handles a specific aspect of the analysis process, making the overall complex system more manageable and maintainable through functional decomposition.
Solution Approach 2:
The patent introduces an intermediary layer of automated analysis between network operations and human operators. This intermediary system processes raw network data, identifies anomalies, and provides structured information to operators, reducing the complexity burden on humans while maintaining system intelligence.
3Measurement precision
If manual root cause analysis is performed, then thorough investigation can be conducted, but human effort and time consumption increase
Solution Approach 1:
The system implements feedback mechanisms where the automated analysis continuously learns from network data patterns and improves its root cause identification accuracy over time. The system analyzes historical data, adjusts its detection algorithms, and refines its understanding of network behaviors to enhance precision without requiring repeated manual investigation.
Solution Approach 2:
The patent performs preliminary automated analysis and data processing before human operators need to intervene. The system pre-processes network data, identifies potential anomalies, and prepares root cause hypotheses in advance, so that when human operators review the information, the heavy lifting of data analysis is already complete, reducing their time effort.
Data Source
Figure 1~2a
Figure 2b~3a
Figure 3b
AI summary
A method for discovering and diagnosing network anomalies. The method includes receiving key performance indicator (KPI) data and alarm data. The method includes extracting features based on samples obtained by discretizing the KPI data and the alarm data. The method includes generating a set of rules based on the features. The method includes identifying a sample as a normal sample or an anomaly sample. In response to identifying the sample as the anomaly sample, the method includes identifying a first rule that corresponds to the sample, wherein the first rule indicates symptoms and root causes of an anomaly included in the sample. The method further includes applying the root causes to derive a root cause explanation of the anomaly and performing a corrective action to resolve the anomaly based on the first rule.