A privacy-preserving data aggregation scheme for dynamic groups in fog computing

Fog computing has garnered significant attention in recent years, since it can bridge the cloud and terminal devices and provide low latency, location awareness, and geo-distribution at the edge of the network. Data aggregation is a prime candidate for fog computing applications. However, most previous works about data aggregation do not focus on the fog computing. In addition, existing secure data aggregation schemes in fog computing usually do not support dynamic groups and arbitrary aggregation functions. In this paper, we construct concrete data encryption, data aggregation and data decryption algorithms, and further propose a privacy-preserving and collusion-resistant data aggregation scheme for dynamic groups in fog computing. Specifically, in the proposed protocol, the cloud server can periodically collect raw data and compute arbitrary aggregation functions on them. Even if some malicious terminal devices collude with the fog device or the cloud server, the honest terminal devices' privacy cannot be breached. The fog device can filter out false data and aggregate all terminal devices' ciphertexts to save the bandwidth. Besides, dynamic join and exit of terminal devices is achieved. Detailed security analysis shows that our scheme holds k-source anonymity. Our scheme is also demonstrated to be efficient via extensive experiments. (C) 2019 Elsevier Inc. All rights reserved.