Blockchain-Based Privacy-Preserving Driver Monitoring for MaaS in the Vehicular IoT
Qinglei Kong, Rongxing Lu, Feng Yin, Shuguang Cui
Abstract
Driving behaviors are highly relevant to automotive statuses and on-board safety, which offer compelling shreds of evidence for mobility as a service (MaaS) providers to develop personalized rental prices and insurance products. However, the direct dissemination of driving behaviors may lead to violations of identity and location privacy. In this paper, our proposed mechanism first achieves the verifiable aggregation and immutable dissemination of performance records by exploiting a blockchain with the proof-of-stake (PoS) consensus. Moreover, to acquire a driver's aggregated performance record from the blockchain, the proposed scheme first realizes quick identification with a Bloom filter and further approaches the target performance record through an oblivious transfer (OT) protocol. A performance evaluation shows that during the acquisition of the records, the computational complexity of our scheme is only related to the scale of the records contained in one transaction. However, the computational complexity of one traditional scheme without a Bloom filter depends on the scale of the records generated during each time slot. Furthermore, the computational complexity of another traditional scheme without aggregation relies on the scale of the records contained in one transaction, as well as the length of a driver's performance history. We also investigate the trade-off between the privacy level and computational complexity, and we determine the optimal number of data records in each transaction.