False data injection attack dataset for classification, identification, and detection for IIoT in Industry 5.0.

Area of Science:

• Cybersecurity
• Industrial Internet of Things (IIoT)
• Industry 5.0

Background:

• The proliferation of Industrial Internet of Things (IIoT) devices in Industry 5.0 introduces significant security vulnerabilities.
• False Data Injection (FDI) attacks pose a critical threat, compromising the availability and operation of connected IIoT devices.

Purpose of the Study:

• To introduce the UKMNCT_IIoT_FDIA dataset for classifying, identifying, and detecting FDI attacks within Industry 5.0 IIoT ecosystems.
• To provide a comprehensive and independent dataset that accurately characterizes diverse IoT environments and network configurations.

Main Methods:

• Development and methodical examination of the UKMNCT_IIoT_FDIA dataset, evaluating its characteristics against real-world IIoT systems.
• Inclusion of varied FDI attack scenarios, encompassing different methods and intensities, to simulate dynamic threat landscapes.
• Demonstration of a multifaceted approach using the dataset for developing and assessing machine learning (ML) and deep learning (DL) algorithms for FDI attack detection.

Main Results:

• The UKMNCT_IIoT_FDIA dataset effectively represents dynamic FDI attack threats in IIoT environments.
• The dataset facilitates the creation and evaluation of robust ML and DL algorithms for efficient FDI attack detection.
• Demonstrated approaches for active attack detection, malicious device identification, and attack categorization show significant effectiveness.

Conclusions:

• The UKMNCT_IIoT_FDIA dataset is a valuable resource for advancing cybersecurity in Industry 5.0 IIoT.
• The proposed methods utilizing the dataset show potential for enhancing the safety and security of IIoT environments.
• Effective detection, identification, and classification of FDI attacks are crucial for securing Industry 5.0 ecosystems.