Root Cause Analysis Using Improved KQI Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring and analyzing service quality in telecommunications networks fail to consistently correlate network anomalies with customer perception, often misidentifying issues due to small degraded samples, which can lead to inaccurate performance assessments.
Innovation Solution
A method that uses improved Key Quality Indicators (KQIs) and hypothesis testing to identify the root cause of anomalous behavior in communication networks by clustering sources based on degradation levels and determining the type of degradation (global or partial) through a learning process and statistical tests like Student's T-test, providing a confidence level in the results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional KQI measurement methods are used to detect network anomalies, then service quality monitoring is performed, but measurement precision deteriorates due to small degraded samples leading to inaccurate performance assessments
Solution Approach 1:
The patent segments the network sources into multiple clusters based on their degradation characteristics. Instead of treating all sources uniformly, the system divides them into groups (clusters) that share similar anomaly patterns. This segmentation allows for more precise measurement within each cluster while maintaining overall system reliability, directly addressing the contradiction between measurement precision and reliability.
Solution Approach 2:
The patent applies partial action by focusing hypothesis testing on specific clusters of sources that exhibit degradation, rather than uniformly analyzing all network sources. By concentrating computational resources and analytical efforts on the most relevant subsets of sources identified through clustering, the system achieves higher measurement precision without compromising overall reliability.
2Measurement precision
If hypothesis testing is applied to determine root causes, then measurement precision improves through statistical analysis, but device complexity increases due to clustering and multiple statistical tests
Solution Approach 1:
The system segments the complex hypothesis testing process into manageable stages: first clustering sources based on degradation characteristics, then applying hypothesis testing to each cluster separately. This segmentation reduces the overall complexity by breaking down the monolithic analysis task into smaller, more tractable sub-tasks while maintaining high measurement precision through systematic statistical evaluation.
Solution Approach 2:
The patent introduces clustering as an intermediary step between raw data collection and hypothesis testing. This intermediary process organizes the data into meaningful groups, making the subsequent hypothesis testing more efficient and interpretable. The clustering acts as a mediator that simplifies the complexity of directly applying statistical tests to all raw data while preserving measurement precision.
3Productivity
If clustering is used to group sources by degradation level, then productivity improves through efficient analysis, but difficulty of detecting and measuring increases due to need for learning processes and threshold determination
Solution Approach 1:
The patent implements self-service through the learning process that automatically determines optimal clustering thresholds and parameters from the data itself. Rather than requiring manual configuration or expert knowledge to set clustering parameters, the system learns these parameters autonomously by analyzing the degradation patterns in the data, thereby improving productivity without significantly increasing the difficulty of detection and measurement.
Solution Approach 2:
The system dynamically adjusts clustering parameters and thresholds based on the learned characteristics of the network data. By changing parameters adaptively rather than using fixed values, the system achieves high productivity in analyzing diverse network conditions while managing the complexity of detection and measurement through data-driven parameter optimization.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure relates to technology for identifying a root cause of anomalous behavior in a communications network. A key quality indicator (KQI) indicative of a performance level associated with a source is received in the communication network. The KQI includes a performance measurement value to identify a performance level of the source having the anomalous behavior. An improved KQI indicative of a level of degradation is calculated based on the anomalous behavior at the source by recovering the KQI to a historical value. Sources of the anomalous behavior are clustered into subsets according to the level of degradation based on the calculated KQI improvement. A global or partial degradation type of the root cause source in the subset having a severe level of degradation is determined using hypothesis testing, and a confidence value is provided for the accepted degradation type.