Network Fault Detection via Voice Quality Parameter Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing network fault detection methods are inadequate as they only detect faults after user complaints, failing to provide real-time detection and timely identification of network issues affecting voice call quality.

Innovation Solution

A method and monitoring center that receive voice streams, acquire source identifiers, codec types, and voice quality parameters to calculate a fault score, determining network faults by comparing baseline values with damage values for parameters like packet loss rate, delay, and noise level, and locating the fault segment within the network.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If network fault detection is performed only after user complaints using test devices, then the detection method is simple, but the fault detection timing is delayed and cannot detect faults in real-time

Engineering Contradiction:
Improvefault detection timingVSAvoiddetection system complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by deploying monitoring agents on network nodes that continuously collect voice quality parameters before faults occur. The system proactively monitors packet loss rate, delay, jitter, and other parameters in real-time, enabling early fault detection without waiting for user complaints or test device interventions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies self-service by enabling network nodes to automatically monitor and report their own voice quality parameters. Each node equipped with a monitoring agent independently collects local voice quality data and reports it to the monitoring center, eliminating the need for external test devices to perform fault detection.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If comprehensive voice quality parameters are collected and analyzed to calculate fault scores, then the fault detection accuracy is improved, but the processing complexity and computational resources increase

Engineering Contradiction:
Improvefault detection accuracyVSAvoidprocessing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the fault detection system into distributed monitoring agents on individual nodes and a central monitoring center. Each agent independently collects local voice quality parameters and calculates local fault scores, then reports to the monitoring center which aggregates results. This segmented approach improves detection accuracy while distributing computational complexity across multiple nodes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses an intermediary approach by introducing a monitoring center that acts as a mediator between distributed monitoring agents and the fault detection decision-making process. The monitoring center collects fault score reports from multiple nodes, performs centralized analysis, and generates comprehensive fault detection results, thereby managing processing complexity centrally while maintaining high detection accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If real-time monitoring of voice streams is implemented across the network, then the fault detection timeliness is improved, but the network bandwidth and system resources are consumed

Engineering Contradiction:
Improvefault detection efficiencyVSAvoidnetwork resource consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent applies the extraction principle by extracting only the essential voice quality parameters (packet loss rate, delay, jitter, etc.) from voice streams for monitoring purposes, rather than processing the entire voice stream. The monitoring agents collect these specific parameters and report them to the monitoring center, enabling real-time fault detection while minimizing network bandwidth consumption by transmitting only critical data.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2741439B1Network failure detecting method and monitoring center
Publication Date: 2018.08.22 HUAWEI TECH CO LTD
  • EP2741439B1 patent drawingFigure 1
  • EP2741439B1 patent drawingFigure 2
  • EP2741439B1 patent drawingFigure 3

AI summary

The present invention discloses a method, a node, and a monitoring center for detecting a network fault and relates to the field of network communications. The method includes: when a node receives a voice stream, receiving a source identifier, a codec type, and voice quality parameters that correspond to the voice stream and are sent by the node; acquiring, according to the codec type and the voice quality parameters, a fault score corresponding to the node; and detecting, according to the fault score corresponding to the node, whether a network fault occurs on a network between the node and a source corresponding to the source identifier. The monitoring center includes: a first receiving module, a first acquiring module, and a first detecting module; and the node includes: a second acquiring module, a third acquiring module, and a second detecting module. In the present invention, a network fault that occurs on a network can be detected in time.