AI Configuration Verifier Layer for Network Security

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current communications network configurations lack an automated mechanism to prevent malicious or destructive configuration commands, even if they are technically and organizationally valid, due to the absence of a second validation layer, which is critical for securing infrastructure against internal threats.

Innovation Solution

A system with a configuration verifier layer using artificial intelligence is introduced between the user layer and network elements, evaluating commands and requiring human authorization for potentially malicious sequences, employing behavioral analysis and sandboxing techniques to assess command impact and prevent unauthorized changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If configurations commands are executed without automated validation, then operational efficiency is maintained, but network security against malicious commands deteriorates

Engineering Contradiction:
Improvenetwork securityVSAvoidconfiguration execution efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary evaluation of configuration commands using AI before they are executed on the production network. The configuration verifier layer analyzes commands in advance, assessing their potential impact and maliciousness, and only allows safe commands to proceed to execution. This resolves the contradiction by maintaining security through pre-validation while preserving efficiency by blocking only harmful commands.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces a configuration verifier layer as an intermediary component between the command input interface and the network elements. This intermediary layer uses AI to evaluate commands, acting as a mediator that filters malicious commands while allowing legitimate ones to pass through. This resolves the contradiction by adding security validation without completely blocking efficient command execution.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If AI evaluation is implemented for all configuration commands, then malicious command detection is improved, but system complexity increases

Engineering Contradiction:
Improvemalicious command detectionVSAvoidsystem architecture complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the network management architecture by inserting a dedicated configuration verifier layer between the user interface and network elements. This segmentation isolates the complex AI evaluation logic into a separate modular component, making the overall system more manageable and maintainable while improving malicious command detection capabilities.

Inventive Principle:
Principle #1Segmentation

3Reliability

If human authorization is required for potentially malicious commands, then security control is improved, but operational time increases

Engineering Contradiction:
Improvesecurity controlVSAvoidcommand execution time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system applies partial human authorization by requiring human approval only for commands that the AI evaluates as potentially malicious or high-risk. Commands that pass AI evaluation as safe are executed automatically without human intervention. This selective approach maintains strong security control for risky operations while minimizing time loss for routine safe commands.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3860079B1Method and system for a secure and valid configuration of network elements
Publication Date: 2024.01.24 DEUTSCHE TELEKOM AG
  • EP3860079B1 patent drawingFigure 1

AI summary

The present invention refers to a system (110) for a secure and valid configuration of network elements (101, 102, 103) of a communications network (100), the system (110) comprising at least one configuration verifier layer (111) which is configured to be installed and sandwiched between a user layer (105) of the communications network (100) and the network elements (101, 102, 103) and which is configured to evaluate configurations commands for the network elements (101, 102, 103), the configurations commands being supplied via the user layer (105), wherein the evaluation is performed using an artificial intelligence (112) which is implemented and executed by at least one computing device which is in operative conjunction with and accessible to the configuration verifier layer (111). The present invention also refers to a respective method and a computer readable medium.