Detecting GPS-spoofing Attack on PMU Data with Phase Angle Unwrapping Technique and Low-Rank Approximation of Hankel Matrix
Imtiaj Khan, Virgilio Centeno · 2022 IEEE Power & Energy Society General Meeting (PESGM) · 2022
Introduction of PMUs to cyber-physical system provides accurate and fast data acquisition, while posing additional risk of being the victim of cyber attack. Both False Data Injection Attack (FDIA) and GPS-Spoofing Attack (GSA) or timing attack can provide malicious data to the cyber system, though these two attacks require different post-attack contingency plans. This difference in contingency plans has engendered the requirement of accurate detection of timing attack and its separation from conventional FDIA. In this article, we have developed a model to address this issue using low-rank approximation of Hankel matrix and phase angle unwrapping technique. Firstly, a method to distinguish timing attack and FDIA using unwrapped phase angle data is developed. Secondly, utilizing low rank Hankel matrix property to differentiate timing attack from FDIA is also presented. Finally, the proposed model is validated by numerical simulation on IEEE 13 node test feeder using simulated GPS-spoofing/timing attack and FDIA at the same time instances. The simulation result confirms the theory. The GSA or timing attack can increase the rank 1 approximation error of Hankel matrix with unwrapped phase angle data by 700 % for 3 sec delay in GPS time-stamp. The rank 1 approximation error increases with the time-delay caused by the GSA. On the other hand, FDIA doesnt show any significant change in the low rank approximation profile of Hankel matrix. Therefore, the proposed method is useful in differentiating GSA from FDIA.