Alarm Correlation Analysis Using Weighted Layered Directed Graph

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network management systems face inefficiencies in analyzing alarm correlations due to unclear recursive relationships, lack of priority presentation, and low analysis speed during alarm storms, as well as challenges in validating alarm analysis apparatuses against predetermined correlation rules.

Innovation Solution

A method and apparatus that generate a weighted layered directed graph and correlation matrix to quickly determine root and derivative alarms, using unidirectional arrows to represent priority relationships and reducing the complexity of recursive relationships, allowing for efficient alarm analysis and validation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If correlation rules are stored one by one manually, then the system can process alarm correlations, but the analysis speed becomes slow with complexity O(MN2)

Engineering Contradiction:
Improvealarm analysis speedVSAvoidcorrelation rule storage complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the correlation rule storage into two parts: direct correlation rules (stored manually) and recursive correlation rules (generated automatically through matrix operations). This segmentation allows the system to handle complex recursive relationships without manually storing every possible rule combination, thereby improving analysis speed while reducing storage complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces the mechanical approach of manually storing and querying each correlation rule with a matrix-based computational system. By using matrix operations to automatically generate recursive correlation rules from direct rules, the system achieves O(MN) complexity instead of O(MN2), significantly improving productivity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manually predefined correlation rules are used, then alarm correlation can be analyzed, but recursive relationships become unclear

Engineering Contradiction:
Improvecorrelation analysis accuracyVSAvoidrecursive relationship clarity
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism where the correlation analysis module queries the correlation rule table and uses matrix operations to generate recursive rules. The system continuously refines the correlation rules by comparing actual alarm relationships with predefined rules, ensuring both reliability and clear representation of recursive relationships through the correlation matrix.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary action by pre-calculating and storing the correlation matrix before alarm analysis. This pre-computation of recursive relationships based on direct rules ensures that when alarms occur, the system can quickly retrieve accurate correlation information without losing recursive relationship clarity during the actual analysis process.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If all alarms are checked one by one during alarm storm, then fault location can be attempted, but the speed of locating and processing fault becomes low

Engineering Contradiction:
Improvefault location accuracyVSAvoidfault processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a universal correlation matrix that can simultaneously analyze multiple alarm relationships. Instead of checking each alarm individually, the matrix-based system processes all alarm correlations in parallel, maintaining fault location accuracy while dramatically reducing the time required during alarm storms.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent applies partial action by focusing the analysis only on alarms that have corresponding rules in the correlation matrix, rather than checking every alarm individually. This selective approach maintains measurement precision for relevant alarms while reducing overall processing time by excluding redundant checks.

Inventive Principle:
Principle #16Partial or excessive action

4Ease of operation

If manually predefined correlation rules are used, then alarm correlation analysis can be performed, but priority relationships among alarms cannot be presented

Engineering Contradiction:
Improvealarm analysis capabilityVSAvoidalarm priority information
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent applies local quality by assigning different priorities to different correlation rules within the same correlation relationship. The correlation matrix stores not only the relationships between alarms but also their priority levels, allowing the system to present both ease of operation and complete priority information simultaneously through differentiated rule weights.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2439877B1Method and device for analyzing alarm correlation, system and method for checking alarm correlation analyzing device
Publication Date: 2018.05.16 ZTE CORP
  • EP2439877B1 patent drawingFigure 1~2
  • EP2439877B1 patent drawingFigure 3~4
  • EP2439877B1 patent drawingFigure 5~6

AI summary

A system for checking an apparatus for analyzing correlation of alarms is provided, and the system (900) comprises: a driving module (904) configured to emulate a network element, construct an alarm in a predetermined alarm correlation rule table, and report the alarm to the apparatus for analyzing the correlation of the alarms; and a checking module (906) configured to check whether the output result of the apparatus for analyzing the correlation of the alarms for the alarm conforms to rules in the alarm correlation rule table. In addition, a method for checking an apparatus for analyzing correlation of alarms is also provided. Meanwhile, a method and an apparatus for determining a correlation matrix of alarms, a method and an apparatus for analyzing correlation of alarms are also provided.