Distributed Voting Mechanism for Network Attack Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Low Power and Lossy Networks (LLNs) face challenges in detecting Denial of Service (DoS) attacks due to limited resources and dynamic conditions, making it difficult to distinguish between legitimate and malicious traffic, especially in distributed DoS attacks.

Innovation Solution

A distributed voting mechanism is implemented where network nodes detect potential attacks using local classifiers and send voting requests to neighboring nodes, which validate the detection using their own classifiers, reducing false positives and conserving network resources by limiting alarms sent to central devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a distributed voting mechanism is implemented for attack detection, then measurement precision and reliability of attack detection are improved, but device complexity and communication overhead increase

Engineering Contradiction:
Improveattack detection accuracyVSAvoiddetection mechanism complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The attack detection system is segmented into multiple independent network nodes, each executing its own local classifier. Instead of a centralized detection system, each node independently analyzes traffic features and generates detection results, which are then combined through a voting mechanism to achieve more accurate and reliable attack detection.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Multiple detection results from different network nodes are merged through a voting mechanism. Each node's classification output is combined with others to determine the final attack detection result, leveraging collective intelligence to improve measurement precision while distributing the computational burden.

Inventive Principle:
Principle #5Merging (Combining)

2Speed

If local classifiers are executed at each network node, then detection speed and responsiveness are improved, but use of energy and processing resources increase

Engineering Contradiction:
Improvedetection speedVSAvoidnode energy consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

Each network node executes a lightweight local classifier that performs partial detection independently. The classifier processes only essential traffic features locally to generate quick detection results, while more complex analysis is distributed across the network through the voting mechanism, achieving fast detection without excessive resource consumption at each node.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If voting requests are sent to multiple neighboring nodes, then reliability of attack detection is improved, but loss of energy and communication overhead increase

Engineering Contradiction:
Improvedetection reliabilityVSAvoidcommunication energy loss
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The voting mechanism acts as an intermediary process that coordinates communication between network nodes. Instead of direct peer-to-peer communication for validation, nodes send voting requests and results through a standardized voting protocol, which manages the communication overhead and energy consumption while maintaining detection reliability.

Inventive Principle:
Principle #24Intermediary (Mediator)

4Measurement precision

If distributed voting is used to validate detections, then false positives are reduced, but loss of time for coordination increases

Engineering Contradiction:
Improvefalse positive rateVSAvoidcoordination time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Network nodes perform preliminary classification using local classifiers before engaging in the voting process. This preliminary action filters out obvious non-attack traffic at each node, reducing the number of voting interactions needed and minimizing coordination time while maintaining low false positive rates through the subsequent voting validation.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3140975B1Distributed voting mechanism for attack detection
Publication Date: 2020.04.15 CISCO TECHNOLOGY INC
  • EP3140975B1 patent drawingFigure 1
  • EP3140975B1 patent drawingFigure 2
  • EP3140975B1 patent drawingFigure 3

AI summary

In one embodiment, a network node receives a voting request from a neighboring node that indicates a potential network attack. The network node determines a set of feature values to be used as input to a classifier based on the voting request. The network node also determines whether the potential network attack is present by using the set of feature values as input to the classifier. The network node further sends a vote to the neighboring node that indicates whether the potential network attack was determined to be present.