Detection and analysis of malicioustraffic using supervised and unsupervised learning

Abstract

Most of the enterprises are transforming their conventional networks into Software Defined Networks (SDN) due to cost efficiency and network flexibility. But recent attacks and security breaches against SDNs expose the security weaknesses of the technology. Distributed Denial of Service (DDoS) is the most common attack launched against various SDN architecture layers. Hence, DDoS has been claimed to be the most hazardous attack and threat to SDNs. The existing mitigation techniques are traffic volumetric methods, entropical methods and traffic flow analysis methods. However, traffic sampling based methods risk incomplete approximation of underlying traffic patterns. The early detection of DDoS attack in the controller is critical and requires highly adaptive and accurate methods. In this work, an effective and accurate DDoS detection method using lion optimization algorithm is proposed. The proposed detection technique is robust enough to detect DDoS attack within the small amount of the attack traffic. Further, to evaluate the performance, the proposed method is compared with the state of the art techniques. newline