Provably Privacy-Preserving Distributed Data Aggregation in Smart Grids

The digitalization of power systems leads to a significant increase of energy consumers and generators with communication capabilities. Using data of such devices allows for a more efficient grid operation, e.g., by improving the balancing of power demand and supply. Fog Computing is a paradigm that enables efficient aggregation and processing of the measurements provided by energy consumers and generators. However, the introduction of these techniques is hindered by missing trust in the data protection, especially for personal-related data such as electric consumption. To resolve this conflict, we propose a privacy-preserving concept for the hierarchical aggregation of distributed data based on additive secret-sharing. To increase the trust towards the system, we model the concept and provide a formal proof of its confidentiality properties. We discuss the attacker models of colluding and non-colluding adversaries on the data flow and show how our scheme mitigates these attacks.