Classification of DDoS traffic for Industrial Internet of Services using Deep learning approaches

The Industrial Internet of Services (IIoS) combines traditional industrial systems with digital technologies, offering numerous advantages like improved efficiency, productivity, and resource optimization. However, the rapid growth of IIoS introduces significant cybersecurity risks. Cyber threats including DDoS attacks, unauthorized access, data breaches, and malware infections, pose a severe risk to IIoS security. Among these threats, DDoS attacks have become a significant concern. DDoS attacks overwhelm IIoS networks with excessive traffic, preventing legitimate users from accessing the network. Such attacks can disrupt IIoS systems, causing downtime and inaccessible services.This study aims to analyze DDoS attacks that target IIoS and explore the effectiveness of deep learning algorithms in detecting DDoS. This research analyzes the performance of four deep learning algorithms, ultimately finding that the DNN and GRU models achieved remarkably high accuracy rates of 99%. This study aims to enhance the ability of Industrial Internet of Services (IIoS) users to identify potential Distributed Denial of Service (DDoS) threats, leading to improved operational security and optimized production processes. By employing the findings of this research, users can effectively detect DDoS hazards, resulting in enhanced productivity and streamlined processes within the IIoS environment.