Exploring a swarm intelligence methodology to identify command and control flow

Botnet poses a significant threat to the Internet today. Reactive techniques that try to detect such an attack and throttle down malicious traffic prevail today but seem not to be very effective. In this paper we present an approach to Botnet detection that is based on the methodology of swarm intelligence. Specifically, particle swarm optimization, a robust stochastic evolutionary algorithm based on the movement and intelligence of swarms, is applied to track the remote controls activities, namely C&C. There exist in literature a few papers in which PSO is used to face the optimization problem. However, no paper exists showing the effectiveness of PSO on this problem. Therefore, PSO is examined in this paper to face the identification of C&C flow. Comparing with other classification techniques, PSO performs a high accuracy.