DDoS Attack Detection using Optimized Back Propagation Neural Network with Artificial Plant Optimization in Cloud Computing

One of the most harmful attacks on cloud computing is distributed denial of service (DDoS). By depleting resources, this attack renders cloud services unavailable to end customers, incurring significant financial and reputational damage. Therefore, creating defenses against this attack is essential for the broad adoption of cloud computing. This paper develops a new detection scheme based on back propagation neural network (BPNN) optimized by artificial plant optimization (APO). The goal of optimization is to improve the BPNN's ability to identify the global optimal value and prevent it from settling for the local minimum. The proposed optimized APO-BPNN detection method uses two benchmark datasets and four different performance measures for analyzing performance in experiments. Experimental results confirmed that the proposed APO-BPNN detection scheme produced high detection accuracy and a faster convergence rate.