Decision Forest Classifier with Flower Search Optimization Algorithm for Efficient Detection of BHP Flooding Attacks in Optical Burst Switching Network

摘要

This research is focused on the efficient classification of BHP flooding attacks in the Optical switching network environment. The burst switching net-work is the backbone of the future generation of the optical network. The burst header packet flooding attacks poses a key security challenge that may have a negative impact on its resource utilization performance and in some cases may create issues like denial of service (DoS). A possible solution to this is to develop efficient classification techniques with optimized features from the network data, so that misbehaving edge notes may be detected at an early stage and remedial action may be taken as a counter measures to protect the network. This research investigates the efficient feature selection by using a novel flower Pollination optimization algorithm (FPA) and then the implementation of a Decision Forest algorithm by Penalizing Attributes (Forest PA) classifier for the detection of flooding attacks. The comparison of the proposed approach with the other existing approaches in terms of various performance metrics such as: Accuracy, precision, recall, sensitivity, specificity and Informedness are presented to understand its suitability.