Active trust evaluation method based on evidence sequence extraction

Abstract

The invention relates to the field of data collection trust and security, in particular to an active trust evaluation method based on evidence sequence extraction. The method can be applied to urban data collection to realize trust evaluation of mobile data collection equipment, and help a system to select a high-quality data provider so as to maximize earnings. According to the method, on one hand, enabling the system to dispatch an unmanned aerial vehicle to access a key monitoring point to obtain sample data, so that active verification trust of a mobile data collector is realized; on the other hand, inferring passive interaction trust of the mobile data collector based on historical communication interactions. And then, weighting the two trust items to obtain comprehensive trust, and labeling and storing the record according to a comprehensive trust value. And finally, extracting a plurality of effective trust evidence sequences, and calculating a normalized trust value based on sequence factors and time factors of the effective trust evidence sequences. After the method is adopted, the system can accurately detect malicious attackers, and safety and data quality are effectively improved.

Description

technical field [0001] The invention relates to the field of trust and security of data collection, in particular to an active trust evaluation method based on evidence sequence extraction. Background technique [0002] At present, data collection has attracted extensive attention in academic research and practical applications, and it constitutes a basic platform for various data-based applications, such as Internet of Vehicles (IoV), smart cities, social networks, e-commerce, etc. Wait. As an open collaborative network, participants in the data collection platform can join the network at will to complete data collection or data relay tasks. Therefore, as the scale of mobile devices connected to the network continues to increase, more and more malicious attackers are entering the network in an attempt to destroy the normal functions of the network, tamper or forge data, and cause serious losses to the data collection platform. In order to solve these problems, researchers...

AI Technical Summary

Problems solved by technology

However, in some existing solutions, trusted components are often directly used for trust calculations, and the influence of untrusted and uncertain components on trust values ​​is ignored.

Or roughly compare the size of each component, and directly classify the trust situation of the evaluator according to the largest component, thus making the trust evaluation result inaccurate

Images