Collusive Attacker Identification Using Binary Behavior Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional trust management schemes in mobile crowdsourcing fail to effectively identify and mitigate collusive attacks by malicious devices that maintain high trust values and collude to manipulate crowd data, leading to incorrect final decisions.
Innovation Solution
An apparatus and method that determine crowdsourcing behaviors of devices based on historical data, measure similarities among devices, and use binary variables to identify collusive attackers by analyzing patterns of incorrect data reporting, employing logic operations and variance analysis to distinguish between collusive and honest devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional trust management schemes are used to evaluate devices, then trust values can be maintained, but collusive attackers can manipulate crowd data and maintain high trust values, leading to incorrect final decisions
Solution Approach 1:
The patent introduces a similarity measurement mechanism as an intermediary between trust evaluation and collusive attack detection. By computing similarity between device behaviors and comparing against thresholds, the system can identify collusive relationships that conventional trust management misses, while preserving the trust value system for honest devices
Solution Approach 2:
The patent replaces the conventional trust management mechanism with a behavior-based similarity analysis system. Instead of relying solely on trust values that can be manipulated, the system uses binary behavior variables and similarity computations to detect collusive patterns, substituting the mechanical trust evaluation with a more sophisticated behavioral analysis
2Measurement precision
If behavior analysis and similarity measurement are performed to identify collusive attackers, then detection accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments device behaviors into discrete binary variables representing specific actions (e.g., reporting correct/incorrect data). This segmentation transforms continuous behavioral data into discrete, computable units that can be efficiently compared using simple logical operations, reducing computational complexity while maintaining detection accuracy
Solution Approach 2:
The patent changes the parameter representation from continuous trust values to binary behavior variables. This parameter transformation simplifies the computational model, allowing similarity measurements to be performed using basic arithmetic and logical operations rather than complex continuous calculations, thereby reducing device complexity
Data Source
AI summary
Embodiments of the present disclosure relate to identifying collusive attackers in mobile crowdsourcing. According to embodiments of the present disclosure, the crowdsourcing behaviors of each device is abstracted as a binary variable and the similarity between any two device is measured, which result in less complex and lightweight in mathematical computation.