A PSO-SVM for Burst Header Packet Flooding Attacks Detection in Optical Burst Switching Networks

An Optical Burst Switching (OBS) network is vulnerable to Header Packet (BHP) flooding attack. In attacks, edge nodes send BHPs at a high rate reserve bandwidth for unrealized data bursts, which leads waste of bandwidth, decrease in performance, and massive loss. Machine learning techniques are utilized detect this attack the OBS network. paper, we propose particle swarm optimization–support vector machine (PSO-SVM) model detecting BHP PSO used optimize parameters SVM. We use dataset provided by UCI warehouse train test model. The experimental results show that detection accuracy PSO-SVM reaches 95.0%, 9.4%, 9.6%, 20.7%, 8% higher than naïve Bayes, SVM, k-nearest neighbor, decision tree. Although DCNN outperforms our model, it requires more processing training time. Collectively, approach effective high-efficiency attacks optical burst switching networks maintaining stability security.