Attack on the Privacy-Preserving Carpooling Service TAROT

The widespread popularity of carpooling services has brought about several privacy concerns, including the collection and use of user location data by service providers. To address these concerns, various carpooling service schemes based on homomorphic encryption have been proposed. TAROT, proposed by Xu et al. (IEEE IOT Journal 2022), aims to be an efficient, accurate, and privacy-preserving carpooling service scheme. In this paper, we show that there is a leakage of location data for users in TAROT. Specifically, we examine a Goldwasser-Micali (homomorphic encryption scheme)-based Equality Determination Algorithm (GMEDA) used in TAROT and propose passive attacks, where honest-but-curious users collude to steal the location information of other users.