Modelagem do Ataque Grayhole ao Protocolo de Comunicação GOOSE usando o Framework ERENO

Abstract

The growing need to enhance cybersecurity in critical infrastructure, specifically in electric substations that communicate via the Generic Object Oriented Substation Event (GOOSE) protocol, calls for effective threat detection and prevention techniques. This protocol, defined by the IEC-61850 standard, protects physical devices by notifying events such as electrical faults. However, its adoption opens gaps for the exploitation of vulnerabilities through attacks whose signatures need to be mapped. In particular, the literature lacks Grayhole attack signatures. This work proposes the modeling and implementation of such attack targeted to the GOOSE protocol. Furthermore, such modeling is incorporated into ERENO, a framework for generating intrusion datasets. The effectiveness of the resulting dataset is validated through five machine learning algorithms, with the J48 algorithm standing out, achieving a 90.68% F1-Score.