Abstract:Massive multidimensional health data collected from Internet of Things (IoT) devices are driving a new era of smart health, and with it come privacy concerns. Privacy-preserving data aggregation (PDA) is a proven solution providing statistics while hiding raw data. However, existing PDA schemes ignore the willingness of data owners to share, so data owners may refuse to share data. To increase their willingness to contribute data, we propose an OPtional dimEnsional pRivacy-preserving data Aggregation scheme (OPERA) to provide data contributors with options on sharing dimensions while keeping their choices and data private. OPERA uses selection vectors to represent the decisions of users and count participants dimensionally and achieves data privacy and utility based on a multisecret sharing method and symmetric homomorphic cryptography. Analyses show that in OPERA, the probability of adversaries breaching privacy is less than 4.68e-97. Performance evaluations demonstrate that OPERA is outstanding in computation and practical in communication.

OPERA: Optional Dimensional Privacy-Preserving Data Aggregation for Smart Healthcare Systems