PP2db: A Privacy-Preserving, P2P-Based Scalable Storage System for Mobile Networks

Abstract

Reputation-based systems that handle millions of users face the problem of simultaneously supporting privacy and trust in an efficient way. In order to scale, often existing systems either sacrifice privacy to preserve trust, or vice versa. The introduction of advanced cryptographic techniques such as Group Signatures might offer a solution, but their applicability to large, distributed systems such as P2P-based ones has yet to be proved. In this paper we introduce PP2db, a privacy-preserving, scalable and distributed storage system targeted at mobile networks, specifically designed to support the anonymous but trusted exchange of Quality of Experience (QoE) information. In such case-study, QoE data is exchanged among users so as to make informed decisions on which network to select at any given time. We demonstrate that by enriching a P2P database with Group Signatures it is possible to create distributed storage mechanisms that guarantee privacy-preserving features, while offering strong trust at the group level. Furthermore, we demonstrate that the resulting architecture can scale in a realistic mobile network scenario to handle millions of users.