Incentivizing Privacy-Preserving Crowdsensing for Smart Transportation

Abstract

We present a novel mechanism to enhance participatory crowdsensing for smart transportation by integrating privacy-preserving data filtering, aggregation and incentive-driven dissemination. Our mechanism filters out inaccurate data before computing the aggregate statistics (e.g., mean, variance) that will be disseminated as incentives to only the users, who contribute useful data to their computations in a privacy-preserving manner. Based on efficient homomorphic cryptosystems, our privacy-preserving mechanism achieves satisfactory data accuracy and privacy simultaneously.