ML-Based Network Fault Resolution for Root Cause Diagnosis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Complex wireless networks face challenges in quickly identifying and resolving system-level experience issues due to the complexity of factors involved, leading to prolonged downtime and impact on users when numerous devices of varying types are involved.

Innovation Solution

A machine learning model is employed to monitor operational parameters and messages within the network, identifying faults and potential root causes, and selecting corrective actions based on probabilities and costs, iterating until the system returns to nominal performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual methods are used to identify root causes in complex wireless networks, then diagnostic accuracy can be maintained, but the time required to resolve faults increases substantially

Engineering Contradiction:

Improveroot cause identification accuracyVSAvoidfault resolution time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent introduces a machine learning model as an intermediary between network operators and complex network diagnostics. The model automatically analyzes network data, identifies root causes, and recommends corrective actions, eliminating the need for manual analysis while maintaining high diagnostic accuracy through trained algorithms.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual diagnostic processes (mechanical human analysis) with an automated machine learning system. The model processes network data, identifies faults, and generates corrective actions automatically, substituting human time and effort with computational processing that operates continuously without fatigue.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If the network system continues operating during fault identification, then service availability is maintained, but system performance degrades due to prolonged downtime

Engineering Contradiction:

Improveservice availabilityVSAvoidsystem performance

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The machine learning model performs preliminary analysis of network data continuously, identifying potential root causes before they escalate into major failures. By detecting issues early and recommending corrective actions in advance, the system maintains performance while preventing severe degradation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a feedback loop where the machine learning model continuously monitors network performance, identifies deviations from normal operation, and recommends corrective actions. This closed-loop system maintains service availability by quickly responding to performance degradation while preventing prolonged downtime through automated monitoring and alerting.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11570038B2Network system fault resolution via a machine learning model

Publication Date: 2023.01.31 JUNIPER NETWORKS INC

AI summary

Disclosed are embodiments for automatically resolving faults in a complex network system. Some embodiments monitor one or more of system operational parameter values and message exchanges between network components. A machine learning model detects a fault in the complex network system, and an action is selected based on a cause of the fault. After the action is applied to the complex network system, additional monitoring is performed to either determine the fault has been resolved or additional actions are to be applied to further resolve the fault.