Securing Smart Grid Digital Twins via Real-World ENF Anchors Against Deepfake Attacks

Integrating digital twins (DTs) into smart grid systems within the Internet of Smart Grid Things (IoSGT) ecosystem introduces significant security challenges. Advanced Deepfake technologies enable malicious actors to fabricate fraudulent DTs, posing substantial risks to the reliability, safety, and integrity of power grids. This paper introduces ANCHOR-Grid, a novel method to authenticate smart grid DTs by using electric network frequency (ENF) signals as anchors in the real world. By capturing ENF variations from physical grid components and embedding this information into their digital counterparts, ANCHOR-Grid establishes a robust mechanism to enhance the security and trustworthiness of virtual representations, mitigating risks posed by Deepfake attacks. To evaluate the effectiveness of ANCHOR-Grid, we conducted comprehensive simulations and experiments within a virtual smart grid environment. We studied both authentic and deepfake DTs of grid components, with the latter attempting to mimic legitimate behavior while lacking the correct ENF signatures. The results demonstrate that ANCHOR-Grid is promising as a robust security layer for smart grid systems within the IoSGT ecosystem. Our work lays the groundwork for future research on the integration of environmental fingerprints into authentication processes for critical infrastructure and validates the importance of considering physical-world cues in securing digital ecosystems.