Optimized Computational Intelligence Driven Decision-Making

3False Data Injection Attack Detection Using Machine Learning in Industrial Internet of Things

Hafizunisa*, Prerna Rai and Damini Sinha

Kalinga Institute of Industrial Technology, Deemed to be University, Bhubaneswar, Odisha, India

Abstract

With the sudden global change and adoption in the field of Internet-of-Things (IoT), it has also made the infrastructure prone to serious cyberattacks. By allowing system autonomy and giving real-time response, the Industrial Internet of Things has aided in resolving numerous industrial issues, but the security challenges for these applications should also be thoroughly explored and taken care of for their successful implementation. There has been a rise in False Data injection (FDI) attacks in cases involving the IIoT. These attacks successfully undermine the decision-making power of the targeted industries by manipulating the critical sensor readings. Recent studies have found that Machine Learning algorithms have the capacity to handle such cyberattacks using their decision-making capabilities and are helpful in detecting a variety of threats arising in the domain of IIoT. In our research, we offer a method for identifying FDI attacks that uses a neural network method called autoencoders (AEs). The AEs helps in finding the deviation between the correlational structure of the corrupted data and the expected learned correlation structure. This is why we suggest using it as a classification strategy to identify FDI attacks. Once the presence ...