An anonymization protocol for continuous and dynamic privacy-preserving data collection

Abstract

Collecting personal data without privacy breaches is important to utilize distributed microdata. Privacy-preserving data collection is anonymizing personal data within the data transmission from data holders to a data collector without privacy breaches. A number of research studies aiming at facilitating the privacy-preserving data collection have been recently conducted. However, the existing studies only allow very particular methods to anonymize data and require too strict assumptions for the private channels between the data holders and the data collector. Thus, these studies suffer from limited data utility and cannot be applied in many environments that does not support the particular requirements. In this paper, we present a novel protocol for the privacy preserving data collection. Unlike existing works, our protocol does not restrict the type of anonymization method and does not require the private channel. Our method requires only the k-anonymity model to prevent privacy attacks, and hence equivalent groups of data holders function as a mechanism for the privacy protection. We further devise a greedy heuristic for dealing with dynamic data holders, and discuss possible attacks on our protocol and prevention of them. Through experiments, we show the performance of the proposed protocol. (C) 2017 Elsevier B.V. All rights reserved.