An identity based security framework for advanced metering infrastructure -

Abstract

The smart grid represents the next generation power grid and will have a substantial role in increasing the reliability and sustainability of energy production and distribution of power delivery networks. Particularly, the integration of information and communication technology is considered to be an evolution in the context of existing power grids. The establishment of two-way communication channels between different components of the smart grid and the utility offers several benefits ranging from early detection of blackouts to time-based rates of energy and demand management programs. Compromising the communication and networking data systems of the smart grid will lead to the risk of compromising reliable and secure power system operations which is the ultimate objective of the smart grid. Therefore, cyber security of the smart grid is of high importance and is considered as one of the highest priorities of the smart grid design. This thesis will focus on the security of the advanced metering infrastructure (AMI) network, which is the last mile of the smart grid, connecting smart meters to the utility. Specifically, the thesis covers topics related to efficient and scalable key management by relying on identity based cryptographic primitives. The first part of the thesis proposes an identity based non-interactive and scalable key distribution system for generating pairwise symmetric keys to be used for link-layer security. In addition, to prevent the discovery and access of the cryptographic keys stored on accessible memory locations physical unclonable functions are adopted. In the second part of the thesis, we present a new lightweight key update and delivery method to securely update the private keys utilized by the identity based cryptosystem. Finally, we propose a new method for deriving keys for the purpose of ensuring a forward secure communication between the meters and the utility at the application level. The proposed method is based on merging identity based cryptography and Diffie-Hellman key exc