Privacy-Preserving Collection of Power Consumption Data for Enhanced AMI Networks

In this paper, we propose a privacy-preserving data collection scheme for enhanced AMI networks. The idea is that each cluster of meters is divided into members and heads. A cluster member should send encrypted subreadings with lifetime values to a number of cluster heads where the lifetime permits the cluster heads to reuse the received subreading for future reporting cycles. Then cluster heads should aggregate all the received/stored subreadings and send the result to a local aggregator which performs a further aggregation process and then send an aggregated reading for the cluster to the utility. If the reading of a cluster member does not change, it should run a countermeasure to traffic analysis to determine whether it needs to send a subreading to one of the cluster heads or not, whereas if the power reading changes or the lifetime of a subreading expires, the cluster member needs to update only one subreading to make the summation of all its subreadings gives the correct reading. In addition, the proposed scheme is more resistive to collusion attacks than existing schemes. Our analysis demonstrates that the proposed scheme can preserve consumers privacy and resist collusion attacks. Our measurements confirm that the proposed scheme can reduce the communication bandwidth by 30%.