A general location-authentication based secure participant recruitment scheme for vehicular crowdsensing

摘要

Vehicular crowdsensing is an emerging paradigm where vehicles perform the sensing task published by the requester so as to earn either monetary or non-monetary incentives. During this process, sensitive information of vehicles (location, identity et al.) are easy to leak. Unfortunately, previous works pay little attention on the privacy-preserving in the participants recruitment process. In this paper, we propose a location authentication based secure participant recruitment (LA-SPR) scheme for vehicular crowdsensing that enables each vehicle to perform (1) location authentication and key transmission simultaneously without any pre-shared secret, and (2) participant recruitment process without revealing the actual location and sensitive information to server or the requester. It's worth mentioning that, our scheme uses a spatial-temporal location tag and fuzzy extractor to realize location authentication and key transmission simultaneously within a round of communication between vehicles and server. We exploit Trust Authority (TA) to collect credible anchor's location tag and vehicle candidates. We regard the sensing score, which is uploaded to the server by each vehicle, as a secret and encrypted with RSA to prevent information leakage. We obtain comprehensive results demonstrating the feasibility, security, and efficiency of our scheme by theoretical analysis and performance evaluation.