A Game-Theoretic Approach for AI-based Botnet Attack Defence
This addresses the lack of assessment tools for evaluating defense strategies against AI-based botnets, though it is incremental as it applies game theory to a specific cybersecurity domain.
The paper tackles the problem of defending against AI-based botnet attacks by proposing a sequential game theory model to analyze attacker and defender strategies, finding through numerical analysis that defense success depends on the number of defense strategies used relative to attack rates.
The new generation of botnets leverages Artificial Intelligent (AI) techniques to conceal the identity of botmasters and the attack intention to avoid detection. Unfortunately, there has not been an existing assessment tool capable of evaluating the effectiveness of existing defense strategies against this kind of AI-based botnet attack. In this paper, we propose a sequential game theory model that is capable to analyse the details of the potential strategies botnet attackers and defenders could use to reach Nash Equilibrium (NE). The utility function is computed under the assumption when the attacker launches the maximum number of DDoS attacks with the minimum attack cost while the defender utilises the maximum number of defense strategies with the minimum defense cost. We conduct a numerical analysis based on a various number of defense strategies involved on different (simulated) cloud-band sizes in relation to different attack success rate values. Our experimental results confirm that the success of defense highly depends on the number of defense strategies used according to careful evaluation of attack rates.