Role-based collaborative information collection model for botnet detection

Abstract

With the growing number of botnet attacks, the botnet detection is becoming increasingly important for the network security. To enhance the existing botnet detection systems which are short of efficient information collection functions, this paper presents a collaborative information collection model with a new 5-tuple structural mode. In the model, we introduce the static and dynamic roles to meet the requirements of information collection and collaboration respectively. Moreover, we give an efficient design for the collection agent and its communication mechanism, which are the core components in the model. Finally, a representative example is given to show that our design for the collection agent can effectively collect the information about the widespread botnet activities, which can help to improve the detection performance and accuracy for a botnet detection system.

Botnet Detection Architecture Based on Heterogeneous Multi-sensor Information Fusion

“…This part adopts a role-based collaborative information collection model, which is our recent work in [32]. This part includes the software and hardware of the information collection system, the main task is to collect all the information related to botnet activities from heterogeneous multi-source sensors.…”