Authenticated down-sampling for privacy-preserving energy usage data sharing

Abstract

Thanks to the availability of fine-grained electricity usage data, electricity customers can benefit from a variety of services from utility companies as well as third party service providers, for energy efficiency, cost saving and incentives, social gaming, and so forth. However, at the same time, advancement of data analytics techniques may jeopardize customers' private, sensitive information. For instance, non-intrusive load monitoring (NILM) techniques could reveal customers' life cycle and style. In general, lowering granularity (sampling frequency) of electricity usage data readings can reduce the amount of information derived. In this direction, we propose a mechanism to allow electricity customers, in order to manage privacy risks, to flexibly down-sample their electricity usage data before sharing it with third-party service providers. On the other hand, our scheme does not invalidate a digital signature on the energy usage data (e.g., one made by utility companies) even after the down-sampling and thus retains verifiability of the data integrity. We then evaluate the overheads in terms of computation and communication and present possible integration into the Green Button information model with minor schema extension.