An Enhanced Approach for Intrusion Detection in Virtual Network of Cloud Computing

Abstract

Cloud computing has offered various advantages in IT businesses by offering everything as a service via Internet. Virtualization and network technologies are the backbone in the construct of cloud infrastructure. However, security of virtual machines against virtual network intrusions is a major concern. To address security problems in cloud, intrusion detection system can play a vital role. We propose an efficient intrusion detection framework for securing the virtual machines in cloud computing against network attacks. It attempts to detect known as well as unknown attacks. For signature based detection, it uses snort, while for anomaly detection, different features selection techniques and the fusion of feasible machine learning techniques are investigated. This framework improves the accuracy of intrusion detection in cloud while reducing the false alerts. The performance and feasibility of the proposed framework is evaluated through performing attacks in real time as well as using a latest intrusion dataset UNSW-NB15.