Advanced EDoS eye: A pragmatic model to mitigate economic denial of sustainability attack in cloud system applying dynamic game-based decision module

Abstract

Low-rate stealthy Distributed Denial of Service (DDoS) attack results by affecting the cloud pricing model is a new form of attack called Economic Denial of Sustainability (EDoS). Existing mitigation models such as EDoS Shield, Controlled Access, and EDoS Eye can partially eliminate EDoS. Firstly, these schemes overlook the attacker’s strategy profile, making their defensive system less resilient. Secondly, using predefined or static threshold values set by some schemes makes the defensive system vulnerable to more false alarms. Thirdly, they use authentication mechanisms like Graphical Turing Test, Crypto Puzzle, or similar graphic-based authentication mechanisms for each new client to identify botnets and filter out malicious traffic. These mechanisms contribute to higher response time. To successfully defend against EDoS attacks, particularly against clever rational attackers, a defender or defensive system needs to choose an optimal defense strategy considering the attacker’s intent. Therefore, we propose Advanced EDoS Eye, a dynamic game theory-based model using non-cooperative, zero-sum multistage game theory in this paper. The proposed analytical Dynamic Game-based Decision Module (D-GBDM) in firewall instance generates dynamic threshold values, substantially reducing the EDoS effect’s payoff in cloud computing. Overall, experimental results favour the proposed model by ensuring the Quality of Service (QoS) of the defense system in the cloud.