Enhanced DDoS Attack Detection in Cyber-Physical Systems Within Supply Chain 4.0 Using Optimized TSO-DBN Deep Learning Technique

It is critical to implement strong detection methods in the developing field of Supply Chain Risk Management 4.0, as cyber-physical systems (CPS) are becoming more vulnerable to Distributed Denial of Service (DDoS) attacks. The authors of this paper present a state-of-the-art approach to detecting distributed denial of service attacks (DDoS) by combining several artificial intelligence (AI) and data processing methods. The first step is Z-score normalization for pre-processing, which involves standardizing the data to make sure that feature scaling is consistent. In order to reduce dimensionality while keeping considerable variation, feature extraction is performed using Principal Component Analysis (PCA). To take the model to the next level, we optimize the feature subset to improve detection performance using Tunicate Swarm Optimization (TSO), which is employed for feature selection. The selected characteristics are then used by a Deep Belief Network (DBN), a model of Deep Learning (DL), to classify anomalies. The DBN uses learnt patterns to differentiate between normal and attack situations. Accuracy, precision, recall, and F1-score are some of the performance indicators used to evaluate the detection system and make sure the model is functional. More than that, the research compares the effects of several feature selection algorithms on the DDoS detection system's overall performance to determine which one is the most effective. In the ever-changing landscape of Supply Chain Risk Management 4.0, this all-encompassing method boosts cyber-physical system resistance to attacks with a 99% success rate and increases DDoS attack detection accuracy and dependability.