Deep Learning-based Secure Machine-to-Machine Communication in Edge-Enabled Industrial IoT
Harsh Mankodiya, Nilesh Kumar Jadav, Sudeep Tanwar, Rajesh Gupta · 2022 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS) · 2022
A still in-progress technology of the Industrial Internet of things (IIoT) involves the use of machine-to-machine (M2M) for communication. It can serve the whole purpose of automation in industries by integrating it with smart systems. However, hindrances in such a system lie in the security area. Traditional systems can be easily compromised by developing competent algorithms. The modern system with robust security sometimes fails to sustain the network traffic and load at deeper layers in the communication systems. To overcome the aforementioned issue, we propose including a trained neural network that can interpret incoming machine data streams in this study. It isolates malicious machines from normal machines by identifying attacks based on machine requests. Hence, this can prevent security breaches early. The performance of the trained neural networks is evaluated considering standard metrics such as F1-score and accuracy scores. The proposed model significantly improves the detection rate of malicious machine requests.