Research On Privacy Protection In Decentralized Social Network

With the rapid development and rise of the Internet,the use of Internet has covered all aspects of life.Social networking applications attract more and more attention of the public.With the rise of mobile terminal technology and wireless technology,mobile Internet came into being,and greatly promote the development of mobile social networks.But in this era of information overload,whether it is social network or mobile social network are facing serious security problems,stronger information security means more broad prospects for development.Most social networks are now centralized architecture in which user information is managed by social network service providers and users do not have control power about their own information,which brought to the user privacy risks.With the emergence of decentralized social network,it is getting more and more public because of the safer architecture compared to the centralized.So we propose the privacy protection programs toward the decentralized social networks.Whether it is centralized or decentralized social network,its privacy and security are mainly related to user information privacy,identity privacy and link privacy.This study originally takes advantage of the inherent nature of the social network and shamir secret sharing protocol to achieve the privacy protection of messages propagation in a decentralized social network.Existing privacy protection schemes for social networks have different privacy protection objectives and different application scenarios,but they all have certain shortcomings.This paper designs a novel privacy protection protocol based on decentralized social network.which can be used to distribute messages to a certain group.In the process of message distribution,it can protect identity privacy,link privacy and guarantee message will not be leaked to the non-target users to protect user information privacy;By means of a server to improve the speed of recover message,a modified-decentralized algorithm appears.We examine the effects to achieve functional goals towards different parameters by doing simulation and emulation experiments and compare the algorithm to other famous privacy protection algorithms recently.The proposed algorithm is compared with the existing privacyprotection algorithms,and the results show that they have more privacy protection and better features than the existing algorithms.The comparison of time required for adding a different level users in the social network shows that the proposed algorithm can better adapt to the changes of the network topology dynamically.By comparing the worst-case computational complexity of the two proposed algorithm,it is shown that the improved algorithm has better time performance without considering the help of a server.In the appendix,we demonstrate and describe the implementation of the privacy protection scheme in the mobile terminal,which show the good application of the privacy protection algorithm.