Private data aggregation with groups for smart grids in a dynamic setting using CRT

Computing the total consumption within a neighbourhood or of a single households in smart grids is important for billing and statistical analysis. Fine granular data used for this purpose, unfortunately, leaks too much privacy sensitive information on the inhabitants and thus raise serious concerns. In this paper, we propose a cryptographic protocol that protects the privacy sensitive measurements while it enables the utility provider to obtain the desired statistical information. Our protocols improves the state-of-the-art in three dimensions. Firstly, from a single execution of the protocol, the utility provider can obtain the total consumption of the whole neighbourhood as well as smaller groups that are created based on their features, e.g. schools, hospitals, etc. Secondly, to the best our knowledge, our protocol is the first one that cope with missing measurements without invoking other protocols or relying on third parties. Thirdly, our protocol relies on simple primitives that can be implemented efficiently even on limited devices, particularly on smart meters. We achieve our goal of having a simple, efficient protocol that is suitable for groups in a dynamic setting by combining the Chinese Remainder Theorem with modified homomorphic encryption. The simplicity and the capabilities of our protocol make it very promising to be deployed in practice as shown in the analysis.