Security and Privacy in Smart Grid
Smart grid utilizes different communication technologies to enhance the reliability and efficiency of the power grid; it allows bi-directional flow of electricity and information, about grid status and customers requirements, among different parties in the grid, i.e., connect generation, distribution, transmission, and consumption subsystems together. Thus, smart grid reduces the power losses and increases the efficiency of electricity generation and distribution. Although smart grid improves the quality of grid's services, it exposes the grid to the cyber security threats that communication networks suffer from in addition to other novel threats because of power grid's nature. For instance, the electricity consumption messages sent from consumers to the utility company via wireless network may be captured, modified, or replayed by adversaries. As a consequent, security and privacy concerns are significant challenges in smart grid.
Smart grid upgrade creates three main communication architectures: The first one is the communication between electricity customers and utility companies via various networks; i.e., home area networks (HANs), building area networks (BANs), and neighbour area networks (NANs), we refer to these networks as customer-side networks in our thesis. The second architecture is the communication between EVs and grid to charge/discharge their batteries via vehicle-to-grid (V2G) connection. The last network is the grid's connection with measurements units that spread all over the grid to monitor its status and send periodic reports to the main control center (CC) for state estimation and bad data detection purposes.
This thesis addresses the security concerns for the three communication architectures. For customer-side networks, the privacy of consumers is the central concern for these networks; also, the transmitted messages integrity and confidentiality should be guaranteed. While the main security concerns for V2G networks are the privacy of vehicle's owners besides the authenticity of participated parties. In the grid's connection with measurements units, integrity attacks, such as false data injection (FDI) attacks, target the measurements' integrity and consequently mislead the main CC to make the wrong decisions for the grid.
The thesis presents two solutions for the security problems in the first architecture; i.e. the customer-side networks. The first proposed solution is security and privacy-preserving scheme in BAN, which is a cluster of HANs. The proposed scheme is based on forecasting the future electricity demand for the whole BAN cluster. Thus, BAN connects to the electricity provider only if the total demand of the cluster is changed. The proposed scheme employs the lattice-based public key NTRU crypto-system to guarantee the confidentiality and authenticity of the exchanged messages and to further reduce the computation and communication load. The security analysis shows that our proposed scheme can achieve the privacy and security requirements. In addition, it efficiently reduces the communication and computation overhead. According to the second solution, it is lightweight privacy-preserving aggregation scheme that permits the smart household appliances to aggregate their readings without involving the connected smart meter. The scheme deploys a lightweight lattice-based homomorphic cryptosystem that depends on simple addition and multiplication operations. Therefore, the proposed scheme guarantees the customers' privacy and message integrity with lightweight overhead.
In addition, the thesis proposes lightweight secure and privacy-preserving V2G connection scheme, in which the power grid assures the confidentiality and integrity of exchanged information during (dis)charging electricity sessions and overcomes EVs' authentication problem. The proposed scheme guarantees the financial profits of the grid and prevents EVs from acting maliciously. Meanwhile, EVs preserve their private information by generating their own pseudonym identities. In addition, the scheme keeps the accountability for the electricity-exchange trade. Furthermore, the proposed scheme provides these security requirements by lightweight overhead; as it diminishes the number of exchanged messages during (dis)charging sessions. Simulation results demonstrate that the proposed scheme significantly reduces the total communication and computation load for V2G connection especially for EVs.
FDI attack, which is one of the severe attacks that threatens the smart grid's efficiency and reliability, inserts fake measurements among the correct ones to mislead CC to make wrong decisions and consequently impact on the grid's performance. In the thesis, we have proposed an FDI attack prevention technique that protects the integrity and availability of the measurements at measurement units and during their transmission to the CC, even with the existence of compromised units. The proposed scheme alleviates the negative impacts of FDI attack on grid's performance. Security analysis and performance evaluation show that our scheme guarantees the integrity and availability of the measurements with lightweight overhead, especially on the restricted-capabilities measurement units.
The proposed schemes are promising solutions for the security and privacy problems of the three main communication networks in smart grid. The novelty of these proposed schemes does not only because they are robust and efficient security solutions, but also due to their lightweight communication and computation overhead, which qualify them to be applicable on limited-capability devices in the grid. So, this work is considered important progress toward more reliable and authentic smart grid.