Privacy-friendly Discovery of Common Friends in P2P Networks

Abstract

In this paper we study the problem of comparing a set of data between two parties in a peer-to-peer network to determine the number of common friends. Several protocols for private set intersection are presented in the literature. When the sets are large these tend to be too slow for many purposes. We consider the problem of two parties finding out how many common friends they have in a privacy preserving way. This problem has arisen in designing a peer-to-peer platform called HELIOS. We present our solution for the problem that is more efficient than older protocols but still sufficiently privacy-friendly for our purposes. The solution is based on iteratively revealing information about the hash values of friends' identities in small increments.