Privacy-Preserving Friend Matching for Mobile Social Networks.

Yaqing Song,Chunxiang Xu,Yuan Zhang,Nan Cheng
DOI: https://doi.org/10.1109/globecom46510.2021.9685386
2021-01-01
Abstract:In this paper, we propose an efficient private set intersection protocol, named LL-PSI, to enable two parties (where each party has an individual set) to obtain the intersection of their sets without leaking other information about their sets to each other. Compared with existing protocols, LL-PSI reduces the computational latency of the intersection between two sets significantly at the expense of communication costs between the parties. Based on LL-PSI, we propose a privacy-preserving friend matching scheme for mobile social networks, dubbed PAIRING. PAIRING allows users to match with those who have common interests while preserving users' private information against the semi-honest server and curious users. We analyze the security of PAIRING and conduct a comprehensive performance evaluation, which demonstrates that PAIRING is secure and efficient.
What problem does this paper attempt to address?