Lightweight privacy-preserving data aggregation scheme for smart grid metering infrastructure protection

Abstract The electric industry's planned shift to smart grid metering infrastructure raised several concerns especially about preserving the privacy. Various data perturbation and aggregation solutions are being developed to address these concerns. The drawback of these solutions is that a simple random noise scheme cannot protect privacy, and there is a need for more advanced perturbation techniques to increase hardware costs of smart metering devices. The proposed data aggregation scheme combines the power of perturbation techniques with crypto-systems in an efficient and lightweight way so that it becomes applicable for devices with limited hardware, such as smart meters. We investigated the privacy preserving capabilities of the proposed aggregation scheme with Holt-Winters and Seasonal Trend Decomposition using Loess prediction methods. The results indicate that the proposed scheme is resilient to both filtering and true value attacks.