Entropy based Defensive Mechanism against DDoS Attack in SDN-Cloud enabled Online Social Networks
Anupama Mishra, Brij Bhooshan Gupta, Dragan Peraković, Shingo Yamaguchi, Ching‐Hsien Hsu · 2021
Online social network (OSN) is now the part of every human life. Users create social relationships with their friends and family members. Today the OSN based on SDN-Cloud becomes the best computation platform for consumers and providers to use scalable and cost-effective services. But at the same time it also has to face many security threats and issues such as privacy, data protection, malwares and virus, sniffing, DDoS, and traffic monitoring. Amongst all threats, Distributed Denial of services (DDoS) is the most severe threat as it utilizes multiple distributed attack sources. We know that every network in the system has an entropy based on their randomness, and increases in the randomness causes entropy to decrease. Therefore, to enhance the detection and mitigation capabilities of a DDoS attack, we use POX controller under the Software Defined Network Cloud OSN. In this paper, we proposed a defensive mechanism to detect such attacks based on the entropy variation of the destination IP address. And then based on entropy value, we block that specific port in SDN-cloud enabled OSN to fight against DDoS threats. We use OpenFlow switches in SDN environment in the Mininet emulator tool. We performed extensive simulation-based experiments and the generated results are used to validate our claims.