The U.S. power grid is being replaced with a smart grid, a complex network of intelligent electronic devices, distributed generators, and dispersed loads, which requires communication networks for management and coordination. Advanced metering infrastructure (AMI) networks are one part of the smart grid to provide two-way communications between smart meters at the consumers' side and the utility companies. AMI networks allow utilities to collect power consumption data at high frequency rates. However, it needs too much communication bandwidth for smart meters to frequently send power consumption data even when the power consumption does not change. Since using cellular networks is one of the best options to AMI networks, the cost of sending this large amount of data is prohibitive. This project considers enhanced AMI networks, where the meters send power consumption data only when there is a significant change. This can significantly reduce the amount of bandwidth needed for sending the power consumption data; however, it creates a new privacy problem. Practical experiment results have confirmed that by observing the data transmission rate and using traffic analysis techniques, the attackers can infer sensitive information about consumers. Therefore, this new privacy problem must be studied and addressed, and strong countermeasures should be developed.
The proposed research systematically combines efforts from privacy, networking, and communication communities. The project promotes a research program designed to: (a) develop schemes for countering traffic analysis in AMI networks by considering different network and adversary models; (b) quantitatively measure the privacy protection provided by the schemes; and (c) evaluate the schemes in a prototype system for validating the proposed research and enabling hands-on experience for both undergraduate and graduate students. The project will significantly contribute to the research on smart grid, as well as computer system security and privacy. The proposed research will lead to a body of knowledge that can be leveraged by the designers of other networks. The proposed project also lends itself to teaching, training and learning of students. A new graduate course focusing on security and privacy aspects of smart grid will be developed. The achievement of the proposed research will be disseminated to academic community and industry via academic conferences and industrial connections.
