Forgery Cyber-Attack Supported by LSTM Neural Network: An Experimental Case Study

Abstract

This work is concerned with the vulnerability of a network industrial control system to cyber-attacks, which is a critical issue nowadays. This is because an attack on a controlled process can damage or destroy it. These attacks use long short-term memory (LSTM) neural networks, which model dynamical processes. This means that the attacker may not know the physical nature of the process; an LSTM network is sufficient to mislead the process operator. Our experimental studies were conducted in an industrial control network containing a magnetic levitation process. The model training, evaluation, and structure selection are described. The chosen LSTM network very well mimicked the considered process. Finally, based on the obtained results, we formulated possible protection methods against the considered types of cyber-attack.

Efficient Cyberattack Detection Methods in Industrial Control Systems

“…The control plant used in the experiments is part of a more extensive laboratory test bed depicted in Figure 1 used by the authors during testing advanced attack methods [ 29 , 30 ]. Three control processes are available: high-speed magnetic levitation, the MPS FESTO workstation, and the thermal heating and cooling stand.…”

Balancing of Low-Voltage Supply Network with a Smart Utility Controller Leveraging Distributed Customer Energy Sources

“…Several techniques exist for the detection and prevention of cyberattacks. In a recent study [31,32], neural networks were explored as a novel approach due to their high level of automation in the learning process, meaning that they require less prior knowledge of the system being safeguarded. Additionally, recommendations suggest the implementation of redundancy in communication protocols, encryption, monitoring for unusual signal behavior through statistical properties and isolating the communication network from external connections (according to [33], the internet is the most commonly used method of cyberattack).…”

A Data-Driven Architecture for Smart Renewable Energy Microgrids in Non-Interconnected Zones: A Colombian Case Study

“…The authors of [27] carried out an exercise on image encryption, since the correct transmission of images was fundamental for their study. Additionally, the authors of [28] used neural networks to create methods for protection against cyberattacks. Moreover, the authors of [29] implemented techniques to improve image authentication.…”