Privacy protection method for trust state evaluation

Abstract

The invention discloses a privacy protection method for trust state evaluation. Majority of the existing solution method is based on the homomorphic encryption and like encryption method, this methodis high in privacy security, but large in computation expenditure, so that the method is hard to operate on a mobile terminal. On the other hand, since the crowd-sourcing sensing policy oriented to the task distribution is continuously perfected, the existing task distribution policy mainly focuses on the expenditure and the completion time, and ignores the privacy of the participant and the influence on the task sensing result by the reliability. Therefore, by matching the basic capacity attribute of the participant in the crowd-souring sensing network with the task attribute, the participantmeeting the basic condition is screened by using a Bloom filter, and then in the candidate participant, the capacity trust of the participant is obtained by further computing the similarity of the capacity weight of the participant and the task weight, and the evaluation on the participant trust value is accomplished by utilizing the relative reputation value of the participant and the task publisher, and the whole participant selection process is realized on this basis. The design method provided by the invention can recruit appropriate participant when the privacy of each of the participantand the requester is protected, and the aim of balancing the privacy protection and the sensing quality is achieved.

Description

technical field [0001] The invention relates to a privacy protection strategy of a crowd-sensing network, in particular to the design of a privacy protection mechanism for trust state-aware participating node selection. Background technique [0002] With the rapid development of mobile Internet technology and applications, mobile intelligent terminal equipment has been widely used and greatly popularized. Smart mobile terminal devices integrate rich sensors and ubiquitous sensing networks, enabling ordinary users to participate in the activities of sensing and collecting surrounding environmental conditions. In this context, the combination of mobile sensing and crowdsourcing ideas has produced new types of Network perception mode, that is, group intelligence perception. At present, crowd sensing has entered a rapid and in-depth development stage, and has penetrated into various aspects such as intelligent transportation travel, infrastructure and municipal management servi...

AI Technical Summary

Problems solved by technology

(3) Task tracking attack: This attack is mostly caused by an honest but curious perception platform, or the perception platform colludes with a malicious attacker. When a participant receives a task from the perception platform, he needs to share some information with the perception platform. Attributes such as time, location, interest, etc. These information alone will not reveal the privacy of participants, but attackers may leak some sensitive attributes of participants by tracking and linking multiple task information

[0005] Aiming at the problem of matching private information in group intelligence task allocation, most of the solutions are based on encryption methods such as exchange encryption and homomorphic encryption. This method has high privacy and security, but it is computationally expensive and difficult to run on mobile terminals.

On the other hand, although the group intelligence strategy for task allocation is constantly improving, most of the existing task allocation strategies focus on the cost and completion time, ignoring the influence of the privacy and reliability of the participants on the task perception results

The separation of task allocation mechanism and trust mechanism makes it difficult for the current task allocation system to operate effectively in an open and dynamic environment

Images