PSI-PPER: A Privacy-Preserving and Efficient Ridesharing Scheme

In recent years, ridesharing has been widely adopted as an instance of sharing economy. When using ridesharing service, users (i.e., drivers and riders) have to share their private trip information with the service provider (SP), which causes great privacy risks to them. In order to protect user’s privacy, based on private set intersection cardinality (short for PSI-CA), we propose a privacy-preserving and efficient ridesharing scheme (short for PSI-PPER). It enables the SP to efficiently match riders with appropriate drivers without learning their private trip information. In PSI-PPER, we adopt “screen and then match” method, which means the SP first screens each rider’s possible matching drivers, and then performs matching operations on each rider and its possible matching drivers. For matching riders and drivers efficiently, the secure ride-matching computation is converted into private set intersection cardinality and is achieved by a one-party to multiparty PSI-CA protocol, which is designed based on vector oblivious linear evaluation and oblivious key-value stores. We prove the security of the proposed scheme theoretically and evaluate its performance through experiments. The results show that the proposed scheme can achieve practical ride-matching accuracy. In addition, PSI-PPER provides higher security and superior efficiency compared to other secure ride-matching schemes with large number of drivers and large trip length.