Security and privacy issues in the smart grid : a decentralized approach

Abstract

Addition of automated components such as smart meters, embedded microprocessors, and the two-ways communication systems from customers to operators to the power grid makes the power grid “smart”. One enhancement that requires the power grid to be “smart” is dynamic pricing. Implementation of dynamic pricing in the smart grid will require frequent transfer of energy consumption data from the consumers to the utilities. Privacy and security issues in transferring this energy consumption data is a widely studied topic. However, almost all of the studies done so far rely on a trusted third party, such as an electrical utilities or a load aggregator, that will have access to all of consumer data. This thesis proposes a Bitcoin-like decentralized model as a solution for secure information transfer within the smart grid, eliminating the presence of a central entity. Hence, a significant portion of this thesis is describes working principles of the Bitcoin network. This thesis specifically focuses on securing bidirectional data transfer between the gateway devices and the electrical utilities. How the information contained in the data is used is left upto the discretion of the consumers. Along with a proposal for a decentralized model, a broad discussion on security and privacy issues in the smart grid as well as cryptographic protocols required for a decentralized smart grid is also presented; a semi-detailed discussion on cryptographic protocols elliptic curve cryptography (ECC) and zero knowledge proof is included. Code for some of the useful cryptographic tools for a decentralized smart grid can be found in the appendix. A novel Bitcoin-like model developed to address security and privacy concerns of consumer data in the smart grid is the contribution of this thesis