Divide & Conquer: A Privacy Safeguarding Framework for the Smart Grid

Smart grid requires granular smart metering data collection at frequent time intervals. However, this introduces unique risks to consumer privacy. To address this challenge, several privacy-preserving data collection frameworks have been proposed. Data aggregation-based frameworks show promise in terms of privacy preservation. However, the related frameworks in the literature either have a high computational overhead on the smart meters or rely on architectures that may lead to single points of compromise. Also, most studies focus on solving the privacy problem and do not consider the integration of accurate billing in the framework. In this paper, we study a multiple aggregator based distributed privacy-preserving framework to address these challenges. The aggregators carry out the bulk of the computational load, thus making the framework lightweight for the smart meters. Our proposed framework and two other related schemes are deployed in an embedded environment for assessing the computational overhead on the smart meters due to temporal aggregation. The schemes are also studied via simulation to assess their end-to-end delay for spatial aggregation and scalability, and it is shown that our scheme outperforms the other two schemes. In addition, we analyze the resilience of our scheme against passive threats to privacy. A qualitative analysis of our scheme against other aggregation-based schemes in the literature is also presented in this paper.