Novel Network IDS in Cloud Computing Based on Optimized Back Propagation Neural Network Using a Self-adaptive Genetic Algorithm

Abstract

Nowadays, Cloud Computing (CC) had become an integral part of IT industry. It represents the maturing of technology and is a pliable, cost-effective platform which provides business/IT services over the Internet. Although there are several benefits of adopting this paradigm, there are some significant hurdles to it and one of them is security. In fact, due to the distributed and open nature of the cloud, resources, applications and data are vulnerable and prone to intrusions that affect confidentiality, availability and integrity of Cloud resources and offered services. Network Intrusion Detection System (NIDS) has become the most commonly used component of computer system security and compliance practices that defends network accessible Cloud resources and services from various kinds of threats and attacks, while maintaining performance and service quality. In this work, in order to detect intrusions in CC environment, we propose a novel anomaly NIDS based on Back Propagation Neural Network (BPNN) classifier optimized using a Self-Adaptive Genetic Algorithm (SAGA). SAGA consists of a standard Genetic Algorithm improved by means of an Adaptive Genetic Algorithm, namely Adaptive Mutation Algorithm. Since, Learning rate and Momentum term are among the most relevant parameters that affect the performance of BPNN classifier, we have employed SAGA to find the optimal values of these two critical parameters, which ensure high detection rate, high accuracy and low false alarm rate. Our novel NIDS is called “ANIDS BPNN-SAGA” (Anomaly NIDS optimized by using Self-Adaptive Genetic Algorithm). The CloudSim simulator and KDD CUP’ 99 dataset are used to verify the proposed system. The obtained experimental results have demonstrated the superiority of the proposed approach in comparison with state-of-the-art methods.