DDoS Attack Detection in Cloud Computing Using Optimized Elman Neural Network Based on Bacterial Colony Optimization and Centroid Opposition-Based Learning

Cloud computing infrastructures are particularly vulnerable to Distributed Denial of Service (DDoS) attacks due to the large-scale and dynamic nature of resources.Large data volumes are handled by cloud settings, which raises the computational cost of detection, and filtering malicious traffic from genuine traffic in such large quantities is difficult.The conventional detection techniques are insufficient.The optimized Elman Neural Network (ENN) used in this study's proposed enhanced DDoS attack detection framework combines centroid opposition-based learning (COBL) with bacterial colony optimization (BCO) called COBCO.The conventional BCO lacks population diversity and can fall into local optima due to random initialization and population update.To overcome the above issues, COBL is used for population initialization and population update to enhance population diversity and avoid local optima issues.By imitating bacterial foraging behavior, the COBCO algorithm improves the ENN's capacity to explore and exploit the solution space, increasing the network's speed of convergence and accuracy of detection.Meanwhile, COBL enhances the learning process by producing a wider range of solid candidate solutions, which offset the drawbacks of conventional opposition-based learning.Extensive simulations show that the suggested strategy outperforms traditional techniques in identifying different kinds of DDoS attacks.