Network Measurement Analysis Using Hierarchical Anomaly Clustering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge in analyzing measurement results from complex systems like communications networks is the vast amount of data, making it difficult to identify the most relevant anomalous results and effectively address system issues.

Innovation Solution

A method involving robust principal component analysis (RPCA) for anomaly detection, followed by hierarchical clustering, is applied to reduce data volume and identify anomalous entities at higher hierarchical levels, enabling targeted network maintenance actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If anomaly detection models are used to analyze measurement results, then anomalous data points can be identified, but the vast amount of data makes it difficult to identify the most relevant anomalous results

Engineering Contradiction:
Improveanomaly detection accuracyVSAvoiddata analysis complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the communication network into hierarchical levels (e.g., core network, access network, radio access network, and specific network elements). Anomaly detection is performed at each hierarchical level, allowing the vast amount of data to be divided into manageable segments. This segmentation reduces the complexity of analyzing the entire dataset while maintaining comprehensive anomaly detection coverage across all network levels.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical dimension to the anomaly detection process. Instead of analyzing all measurement results in a single flat space, the system organizes data into multiple hierarchical levels and performs anomaly detection across these dimensions. This dimensional transformation allows the system to manage vast amounts of data more effectively by distributing analysis across hierarchical layers, thereby reducing overall analytical complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If comprehensive measurement results are analyzed, then system issues can be detected, but the large data volume reduces analysis efficiency

Engineering Contradiction:
Improvesystem monitoring reliabilityVSAvoiddata analysis productivity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent divides comprehensive measurement results into hierarchical segments, allowing anomaly detection to be performed on smaller subsets of data at each level. This segmentation maintains comprehensive monitoring coverage while significantly improving analysis productivity by avoiding the need to process the entire dataset simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary anomaly detection at lower hierarchical levels before analyzing higher levels. This preliminary action filters out many normal data points early in the process, reducing the volume of data that requires intensive analysis at higher levels. As a result, the system maintains high reliability through comprehensive analysis while improving productivity by reducing the computational burden on higher-level analysis.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4356313B1Analyzing measurement results of a communications network
Publication Date: 2026.03.25 ELISA OYJ
  • EP4356313B1 patent drawingFigure 1~2
  • EP4356313B1 patent drawingFigure 3
  • EP4356313B1 patent drawingFigure 4

AI summary

Analyzing measurement results of a target system. Measurement results are obtained (401). The measurement results include multiple data entries, and each data entry includes multiple data values on a lowest hierarchy level and hierarchy information defining with which entities the entry is related to on different hierarchy levels. An aggregated anomaly score is determined (402) for each data entry. Data entries, wherein the aggregated anomaly score fulfils predefined criteria, are chosen for further analysis. Hierarchical clustering is performed (404) on the chosen entries based on dissimilarity of the chosen entries to combine at least some of the chosen entries together; and the hierarchically clustered entries are used (405) to identify one or more anomalous entities on hierarchy levels above the lowest hierarchy level.