Deep Learning for Attack Detection in Industrial IoT Edge Devices

In recent times, system controllers are being fully deployed via the Industrial Internet of Things (IIoT), which significantly enhances the economy and manufacturing industry. Digital security concerns are also brought on by this evolution unfortunately. Due to the fact that a large portion of the value of IIoT systems are located at the edge level, attackers may find these as attractive targets. So, it is crucial to monitor edge system components and spot harmful activity using an effective diagnostic model in order to protect them. This study suggests a deep learning-based attack detection model that can be trained and tested using data gathered from a gas pipeline system. Improved Random Neural Network and Long Short Term Memory Networks are incorporated for the attack detection purposes. The proposed model achieves an accuracy of 97.8% and outperforms the other existing detection models.