BPRT: A blockchain‐based privacy‐preserving transaction scheme based on an efficient broadcast encryption with personalized messages

Abstract

As blockchain has drawn more and more attention, much research has developed different transaction schemes with privacy preserving specialized in various fields. The nature of blockchain makes the transaction information public, which presents much risk about the leakage of user privacy. It is quite necessary to design a flexible scheme to provide privacy preserving of transaction contents with reliable audit in blockchain that is lacked in the previous works. In this paper, we propose a blockchain‐based privacy‐preserving transaction scheme based on an efficient broadcast encryption with personalized messages called BPRT. First, we propose a fully anonymous broadcast encryption with personalized messages (BEPM). Compared with existing works, it is fully anonymous, and it has lower computation cost without pairing operations. It also avoids the key escrow problem existing in other BEPM schemes. Then, we construct the BPRT based on our BEPM. Our BPRT ensures the privacy preserving of transaction contents in group transmissions, and the encrypted transaction contents avoid exposing user identity information at the same time. It enables auditor to moderately and reliably make audit of transaction contents. Furthermore, we present formal security analysis of the proposed scheme. It has been proved that our scheme satisfies transaction confidentiality, receiver anonymity and audit reliability. Subsequently, experimental results indicate that the proposed BEPM enables fast encryption and decryption, and the computation of our BPRT is in milliseconds.

A privacy‐sensitive data identification model in online social networks

“…Although these methods could infer the privacy of users with high accuracy by fully exploiting the observable attributes as well as relationships in OSNs, it is still unclear which data would play the key roles in privacy inference and how privacy-sensitive data could be identified. On the other hand, although some studies 10,11,23 mentioned the need of identifying data that could affect the privacy of users in privacy inference, there was no method that could be used to conduct overall evaluation on different dimensions of data in the complex network. To identify privacy-sensitive data about OSN users, we do not simply pursue the results of privacy inference as the ultimate goal.…”