Differentially Private Smart Metering: Implementation, Analytics, and Billing

Smart power grids offer to revolutionize power distribution by sharing granular power usage data, though this same data sharing can reveal a great deal about users, and there are serious privacy concerns for customers. In this paper, we address these concerns using differential privacy. Differential privacy is a statistical notion of privacy that adds noise to provide privacy guarantees. One privacy threat is the aggregation of time series data, and we therefore apply a trajectory-level form of differential privacy to guard against such privacy threats. In particular, we consider input-perturbation privacy, which adds noise directly to sensitive data streams before sharing them. We apply it in this work to provide privacy guarantees on an individual basis. We then address the impact of privacy upon two key grid stakeholders: the utility and the accuracy of its analytics of interest, as well as customers and the financial impact upon their utility bills. Both impacts are shown to be modest, even with strong privacy guarantees. Simulation results are provided using actual power usage data, demonstrating the viability of this approach in practice.