Decentralized Privacy-Preserving Reputation Management for Mobile Crowdsensing

In mobile crowdsensing, mobile devices can be fully utilized to complete various sensing tasks without deploying thousands of static sensors. This property makes that mobile crowdsensing has been adopted by a wide range of practical applications. Since most crowdsensing platforms are open for registration, it is very possible that some participants might be motivated by financial interest or compromised by hackers to provide falsified sensing data. Further, the urgent privacy-preserving need in this scenario has brought more difficulty to deal with these malicious participants. Even though there have existed some approaches to tackle to problem of falsified sensing data while preserving the participants’ privacy, these approaches rely on a centralized entity which is easy to be the bottleneck of the security of the whole system. Hence in this paper, we propose a decentralized privacy-preserving management scheme to address the problem above. At first, the system model is present based on the consortium blockchain. Then, a novel metric to evaluate the reliability degree of the sensing data efficiently and privately is designed by leveraging the Paillier crytosystem. Based on this metric, how to update reputation values is given. Extensive experiments verify the effectiveness and efficiency of the proposed scheme.